DevelopmentofControltechniquesforDirectAC‐ACMatrixConverterfedMulti‐
phaseMulti‐motorDriveSystem
Mohammed Ahmed Saleh
Electrical & Electronics Engineering College of Engineering & Science
Victoria University, Melbourne, Australia
Submitted in fulfilment of the requirements of the degree of PhD
September 2013
i
ABSTRACT
Therearenumerousindustrialapplications,suchaspapermills,locomotivetraction,oiland gas,mining andmachine tools, which require high performance control ofmorethan one electric motor simultaneously. When more than one electric motors areemployedinanelectricdrive,itiscalled,‘multi‐motordrive’.Thesemulti‐motordrivesare generally available in two configurations. The first one consists of a number ofthree‐phasevoltagesource invertersconnected inparallel toa commonDC link,eachinverterfeedingathree‐phaseACmotor.Thisconfigurationallowsindependentcontrolof all machines by means of their own three‐phase voltage source inverters (VSIs).Nevertheless, this configuration needs n number of voltage source inverters forsupplyingnnumberofACmachines.Thesecondconfigurationcomprisesoneinverter,which feeds multiple parallel‐connected three‐phase motors. However, the laterconfigurationdoesnotallowindependentcontrolofeachmotorandissuitableonlyfortraction application. The power converter supplying the drive system, areconventionally,voltagesourceinverters.However,alternativesolutioncouldbeadirectAC‐AC converter that can supply the electric drive system. Exploring this alternativesolutionisthesubjectofthisthesis.Thus fordecoupleddynamiccontrolofACmachinesworking inagroup(multi‐motordrive) is possible by employingmulti‐phase (more than three‐phase) motors, wheretheirstatorwindingsareconnectedineitherseriesorinparallelandthecombinationissuppliedfromasinglemulti‐phasepowerconverter.This thesis explores the control techniquesofmulti‐phasedirectAC‐AC converter forsuch specific series and parallel‐connected multi‐phase motor drives. The researchpresentedhereutilisesadditionaldegreesoffreedomavailableinamulti‐phasesystemtocontrolanumberofmachinesindependently. Theconceptisbasedonthefactthatindependent flux and torque control of anyACmachine, regardless of the number ofstatorphasesrequirescontrolofonlytwostatorcurrentcomponents. This leavestheremainingcurrentcomponentsfreetocontrolothermachineswithinthegroup.The multi‐phase multi‐motor drive system fed using multi‐phase direct AC‐ACconverter need precise Pulse Width Modulation (PWM) technique to independentlycontrolthedrivesystem.ThesubjectofthisresearchistoproposePWMtechniquesforsuch configurations. The thesis focuses on four different cases; five‐phase, six‐phase(symmetrical and asymmetrical), and seven‐phase system. Five‐phase and six‐phasedrive systems consists of two motors, and seven‐phase drive system controls threemotors. The thesis presents various PWM techniques aimed at these driveconfiguration. Carrier‐based, carrier‐based with harmonic injection and direct dutyratiobasedPWMtechniquesarepresentedinthethesis.Theindependenceofcontrolofvariousmotorsareshownbysimulationandexperimentation.Although,theproposedtechniques are equally applicable to series‐connected drives and parallel‐connecteddrives,thethesisfocusesontheformerdriveconfiguration.Analytical,simulationandexperimentalapproachisusedthroughoutthethesis.
ii
Declaration
“I, Mohammed Ahmed Saleh, declare that the PhD thesis entitled “Development of Control techniques for Direct AC-AC Matrix Converter fed Multi-phase Multi-motor Drive System” is no more than 100,000 words in length including quotes and exclusive of tables, figures, appendices, bibliography, references and footnotes. This thesis contains no material that has been submitted previously, in whole or in part, for the award of any other academic degree or diploma. Except where otherwise indicated, this thesis is my own work”.
Signature Date
iii
ACKNOWLEDGEMENT
The research work presented in this thesis has been carried out under the invaluable
and expert guidance of Professor Akhtar Kalam, Dr. Atif Iqbal and Prof. Haitham Abu-Rub. I
consider it a privilege to have been associated with them for carrying out the research work
towards my PhD degree. I have greatly benefited from their deep insight into the subject and
constant support during numerous fruitful discussions. It is with a deep sense of my affection
and gratitude that I wish to express my sincere thanks to them.
It is my duty to thanks the Head of the Departments, Department of Electrical
Engineering, Qatar University and Department of Electrical & Computer Engineering, Texas
A&M University at Qatar for their help in completing the experimental work at their
laboratory.
My sincere thanks are due to Mr. SK. Moin Ahmed, Research Associate, Department
of Electrical & Computer Engineering, Texas A&M University at Qatar, Qatar, for his
endless support in commissioning the experimental set up used in this work.
I must also acknowledge here the fruitful discussions that I had time and again with
my colleagues, Mr. Moidu Thavot, Mr. Khaleequr Rahman and Mr. Ahmad Anad of Qatar
University, Doha, Qatar.
I also sincerely thanks Qatar Foundation for providing fund for the research work
through the NPRP 4-152-02-053.
I would like to thank my parents, my wife’s parents and my relatives for their moral
support for pursuing this work. Without their constant assurances and assistance, completion
of this project would have not been possible.
Finally, the patience, understanding and co-operation provided throughout the work
by my wife Mona and children Abdullah, Reema and Nour, who had to bear the constraints
of my long working hours during the last three years, are greatly appreciated.
iv
List of Publications
1. Iqbal, A., Payami, S., Saleh. M., Ahmad, A.A., Moinuddin, S., (2013), “Five-phase AC/DC/AC converter with PWM rectifier”, Australian journal of Electrical & Electronics Engineering (Accepted).
2. Saleh, M,Iqbal, A., Ahmed, Sk. M., Abu-Rub. H., Kalam, A., (2013) “Direct Duty ratio based PWM control of a five-phase Matrix Converter supplying five-phase two-motor drive system”, Australian Journal of Electrical & Electronics Engg., vol. 9, issue 3, pp. 283-293, Feb. 2013
3. Ahmed, SK.M., Iqbal, A., Abu-Rub, H., Rodriguez, J., Saleh, M., (2011), “Simple carrier-based PWM technique for a three to nine-phase Matrix Converter”, IEEE Trans. On Ind. Elect., vol. 58, no. 11, pp. 5014-5023, Nov. 2011.
4. Iqbal, A., Saleh, M., Ahmed, Sk. M., Abu-Rub. H., Kalam, A., (2013) “Simple carrier-based PWM technique for a three to quasi six-phase Matrix Converter”, Australian Journal of Electrical & Electronics Engg., vol. 9, issue 3, pp. 295-304, Feb. 2013.
5. Ahmed, M. SK, Iqbal A., Abu-Rub, H., Saleh, M., Kalam. A., (2013), “Vector Control of a Five-phase induction motor supplied by a Non-square 3X5 phase Matrix Converter”, Australian Journal of Electrical Engineering, vol. 10, no. 1, March 2013, pp. 55-63.
6. Saleh, M., Iqbal, A., Thavot, M., Kalam, A., (2013), “Modeling of seven-phase series-connected three-motor drive system”, 7th IEEE GCC COnf., Doha, Qatar, Nov. 2013 (review).
7. Saleh, M., Khan, M.A., Iqbal, A., Moin, SK., Kalam, A., (2013) “Three to seven phase Matrix Converter supplied seven-phase three-motor drive”, Journal of Innovation in Electronics and Communication, vol. 2, Issue 3, pp. 6-13. July-Dec. 2013.
8. Ahmad, R., Iqbal, A., Saleh, M., Kalam, A. (2013), “Performance Analysis of Soft Starter Based Control of Five-phase Induction Motor”, 2013 IEEE PES General Meeting, 21 - 25 July 2013, Vancouver, BC, Canada. CD-ROM paper.
9. Islam, S., Bakhs, Ilahi, Iqbal, A., Saleh, M., Kalam, A. (2013), “Stability Analysis of a Series-Connected Five-phase Induction Motor Drive System using Flux-linkage model” Paper 4604, 8th IEEE Int. conf. on Industrail electronics and applications, 19-21 June, 2013, Melbourne, Australia, CD-ROM paper.
10. Iqbal, A., Pyami, S., Saleh, M., Anad, A., Kalam, A., (2012), “Analysis of Five-phase AC/DC/AC active front end converter”, Australian Universities Power Engineering Conf (AUPEC), 26-29 Sept., Bali, Indonesia, CD-ROM paper.
11. Saleh, M.,Iqbal, A., Moin, SK., Kalam, A., (2012), “Matrix Converter based Five-phase series-connected induction motor drive”, Australian Universities Power Engineering Conf (AUPEC), 26-29 Sept., Bali, Indonesia CD-ROM paper.
12. Saleh. M., Iqbal, A., Ahmed, SK.M, Abu-Rub, H., Akhtar, K., (2011), “Carrier-based PWM technique for a three to six phase Matrix Converter supplying six-phase two-motor drives”, IEEE IECON, 7-10 Nov, Melbourne, Australia pp. 3347-3352.
v
Table of Contents
ABSTRACT ............................................................................................................................ i
LIST OF FIGURES ............................................................................................................... ix
LIST OF TABLES ............................................................................................................... xii
LIST OF ABBREVIATIONS ............................................................................................ xiii
LIST OF SYMBOLS .......................................................................................................... xiv
Chapter 1 Introduction ............................................................................................................... 1
1.1 Preliminary Remark .................................................................................................... 1
1.2 Power Electronic Converters ....................................................................................... 2
1.3 Features of Multi-phase Motor Drive .......................................................................... 5
1.4 Analyzed Drive Topologies ........................................................................................ 6
1.5 Research Objectives .................................................................................................. 12
1.6 Layout of the Thesis .................................................................................................. 12
1.7 Novelty and Contribution of Thesis .......................................................................... 14
References: ........................................................................................................................... 15
Chapter 2 Literature Review .................................................................................................... 17
2.1 Introduction ............................................................................................................... 17
2.2 Multi-phase Motor Drive Systems ............................................................................ 18
2.3 Three-phase input and three-phase output Matrix Converter .................................... 22
2.4 Multi-phase Matrix Converter ................................................................................... 23
2.5 Summary ................................................................................................................... 28
References: ........................................................................................................................... 29
Chapter 3 Over-View of Modelling and Control of three-phase by three-phase Matrix-Converter.................................................................................................................................. 33
3.1 Introduction ............................................................................................................... 33
3.2 Control Algorithms for matrix converter .................................................................. 33
3.3 MODULATION DUTY CYCLE MATRIXAPPROACH ....................................... 34 A. Alesina-Venturini 1981 (AV method) ......................................................................... 36 B. Alesina-Venturini 1989 (Optimum AV method) ......................................................... 37
3.4 Space Vector Control Approach ............................................................................... 38
vi
A. SVM Technique ..................................................................................................... 38
3.7 Carrier-Based PWM Approach ................................................................................. 44 A. Carrier Based Control Strategy .................................................................................... 44 B. Application of Offset Duty Ratios ............................................................................... 45
3.8 Direct Duty Ratio Based PWM approach ................................................................. 47
3.9 Summary ................................................................................................................... 52
References: ........................................................................................................................... 52
Chapter 4 Modelling of Multi-phase Multi-motor Drive System ............................................ 54
4.1 Introduction ............................................................................................................... 54
4.2 Modelling of five-phase series-connected two-motor drives .................................... 56 A. Modelling of series-connected five-phase two-motor drive ................................. 56 4.2b Model in the Rotating Reference Frame ................................................................ 60 4.2c Model in the stationary reference frame ................................................................ 63 4.2d Simulation Results ............................................................................................. 65
4.3 Modeling of a Six-phase Series-connected Two-motor Drive System ..................... 66 4.3a Phase Variable Model ............................................................................................ 67 4.3b Model in the Rotating reference frame .................................................................. 70 4.3c Transformation of the model into the stationary common reference frame ......... 72
4.4 Modelling of A Seven-Phase Series-connected Three-Motor Drive System ............ 74 4.4a Phase Variable Model ............................................................................................ 76
4.4b Model in the rotating reference frame ............................................................... 83
4.4 c Transformation of model in to the stationary common reference frame ........... 91
Torque equation of the machine become ............................................................................. 93 4.4d Simulation Result .................................................................................................. 95
4.5 Summary ................................................................................................................. 101
References .......................................................................................................................... 101
Chapter 5 Space Vector Modelling of Multi-phase Matrix Converter .................................. 104
5.1 Introduction ............................................................................................................. 104
5.2 Space Vector Model of a Three-phase to Five-phase Matrix Converter ................. 104
5.3 Space Vector Model of Three-phase to Six-phase Direct Matrix Converter .......... 127
5.4 Space Vector Model of Three-phase to Seven-phase Direct Matrix Converter ...... 136
5.5 Summary ................................................................................................................. 144
References: ......................................................................................................................... 145
Chapter 6 Carrier Based PWM Schemes for Feeding Multi-Motor Drive System ............... 146
vii
6.1 Introduction ............................................................................................................. 146
6.2 Generalized Carrier-based PWM Techniques for three-to N-phase ....................... 147 6.2a Application of offset duty ratios ............................................................................. 150 6.2b Without Common Mode Voltage Addition ............................................................ 152 6.2.c With Common Mode Voltage Addition ................................................................ 152 6.2d Output Side over Modulation ................................................................................. 153 6.2e Modulator Gain.................................................................................................... 153
6.3.1 Carrier Based PWM Technique for a Three-to-Five Phase Matrix Converter .... 153 6.3a Three-To-Five Phase Matrix Converter .................................................................. 154 6.3b Simulation Results .................................................................................................. 156
6.3.2 Career Based PWM Technique for a Three-to-Five Phase Matrix Converter for Supplying Five-phase Two-motor Drives .......................................................................... 157
6.4a Five-Phase Two-Motor Drive System .................................................................... 158 6.4b Carrier Based PWM Technique for a two motor system ........................................ 160 6.4c Simulation Results for RL Load ............................................................................. 162 6.4d Simulation Results for Motor Load .................................................................... 166
6.5 Carrier Based PWM Technique for a Three-to-Six Phase Matrix Converter ......... 170 6.5a Three-to-Six Phase Matrix Converter system ......................................................... 171 6.5b Application of Offset Duty Ratio ........................................................................... 172 6.5c With Harmonic Injection ..................................................................................... 173 6.5d Output Voltage Magnitude .................................................................................. 175 6.5e Simulation Results .................................................................................................. 175
6.6 Carrier Based PWM Technique for a Three-to-Six Phase Matrix Converter for Supplying Six-phase Two-motor Drives ............................................................................ 177
6.6a Six-Phase Series-Connected Two-Motor Drive Configuration .............................. 178 6.6b Carrier-Based PWM Technique For Six-Phase Two-Motor Drive ........................ 179 6.6c Application of Offset Duty Ratio............................................................................ 180 6.6d Simulation Results .................................................................................................. 182
6.7 Carrier Based PWM Technique Strategies for a Three-to-Seven Phase Matrix Converter ............................................................................................................................ 186
6.7a Application of Offset Duty Ratio............................................................................ 187 6.7b Without Common-mode voltage addition .............................................................. 189 6.7b With Common mode voltage Addition .................................................................. 189 6.7c Simulation Results .................................................................................................. 190
6.8 Seven-Phase Series Connected Three-Motor Drive System ................................... 193 6.8a Carrier-Based Pulse Width Modulation Technique ................................................ 194 6.8b Simulation Results .................................................................................................. 197 6.8c Independent Control at Identical Frequencies ........................................................ 198 6.8d Independent Control at Three Different Frequencies ............................................. 201
6.9 Experimental Investigation ..................................................................................... 203 6.9a Experimental Set-up ............................................................................................... 203
viii
6.9b Experimental Investigation of two motor supply .................................................. 211 6.9c Experimental investigation on 3x6 phase Matrix Converter .................................. 212
6.10 Summary ................................................................................................................. 215
References .......................................................................................................................... 215
Chapter 7 Direct Duty Ratio Based Pulse Width Modulation of Multi-phase Matrix Converter................................................................................................................................................ 217
7.1 Introduction ............................................................................................................. 217
7.2 Direct Duty Ratio based PWM Technique for a Three-phase to n-phase Matrix Converter for Single-motor drive system ........................................................................... 218
Case-I: ................................................................................................................................ 218
7.3 Direct Duty Ratio Based PWM for 3-phase to 5-phase Matrix Converter ............. 223
7.4 Simulation Results for Five-phase Single-motor Drive .......................................... 226
7.4 DDPWM of three-to-fivephase Matrix Converters for five-phase two-motor drive 237
7.5 Simulation Results for Two-motor drive system .................................................... 240
7.7 Summary ................................................................................................................. 247
References .......................................................................................................................... 247
Chapter 8 Conclusions and Future Work ............................................................................... 248
8.1 Conclusions ............................................................................................................. 248 8.2 Future Work ......................................................................................................... 252
ix
LIST OF FIGURES
Fig. 1.1 General block diagram of a Variable speed drive 2 Fig. 1.2. Possible discrete implementations of a bi-directional switch. 3 Fig. 1.3. Three-phase multi-motor drive system. 7 Fig. 1.4a. Five-phase series-connected two-motor drive structure. 9 Fig. 1.5. Outlay of a three-phase to n-phase direct Matrix Converter. 10 Fig. 2.1. Power Circuit topology of Three-phase to k-phase Matrix Converter. 24 Fig. 2.2. Single-sided Matrix Converter with unidirectional power switches as shown in
reference [2.58]. 27 Fig. 2.3. Indirect Matrix Converter with rectifier and inversion stage as shown in reference
[2.58]. 27 Fig. 3.1. Three phase (input) to single phase (output) basic connection. 34 Fig. 3.2. General topology of three phase to three phase Matrix Converter. 35 Fig. 3.3. Double-sided switching pattern in a cycle period Tp. 37 Fig. 3.4. Direction of the output line-to neutral voltage vectors generated by the active
configurations. 40 Fig. 3.5. Double-sided switching pattern in a cycle period Tp [3.7]. 44 Fig. 3.6. Modified offset duty ratios for all input phases. 46 Fig. 3.7. Output and Switching pattern for kth phase in the Case I. 50 Fig. 3.8. Output and Switching pattern for kth phase in the Case II. 52 Fig. 4.1. Five-phase series-connected two-motor drive structure. 56 Fig. 4.2. Response of Five-phase Two-motor drive supplied by ideal voltage source. 66 Fig. 4.3. Connection diagram for series connection of a six-phase and a three-phase machine.
67 Fig. 4.4. Response of Six-phase Two-motor drive supplied by ideal voltage source. 74 Fig. 4.5. Seven Phase Series-Connected Three-Motor system. 75 Fig. 4.6. Response of Seven-phase Three-motor drive supplied by ideal voltage source. 97 Fig. 4.7, Torque and speed characteristics of Machine 1 97 Fig. 4.8. Torque and Speed characteristics of Machine 2. 98 Fig. 4.9. Speed and Torque Characteristics of Machine 3. 98 Fig. 4.10. Current ‘Iα-Iβ’. 98 Fig. 4.11. Current ‘Ix1-Iy1’. 99 Fig. 4.12. Source current ‘Ix2-Iy2’. 99 Fig. 4.13. Spectrum of source phase ‘a’ voltage. 99 Fig. 4.14. Spectrum of source voltage Vα. 100 Fig. 4.15. Spectrum of source voltage Vx1. 100 Fig. 4.16. Spectrum of source voltage Vx2. 100 Fig. 5.1. Basic topology of a three-phase to five-phase Matrix Converter. 105 Fig. 5.2. Switching states of Group-1. 108 Fig. 5.3. Switching states of Group-2 [4,1,0]. 110 Fig. 5.4. Switching states for Group-4 [3,2,0]. 116
x
Fig. 5.5. Output Voltage space vectors corresponding to the permitted switching combinations. 118
Fig. 5.6. Output large length voltage space vectors. 119 Fig. 5.7. Output medium length voltage space vectors. 122 Fig. 5.8. Output small length voltage space vectors. 124 Fig. 5.9. Three-six phase direct Matrix Converter. 127 Fig. 5.10. Three-phase to six-phase DMC outpour voltage (Vdq). 134 Fig. 5.11. Three-phase to seven-phase direct Matrix Converter. 136 Fig. 5.12. Adjacent line space vectors in d-q plane. 141 Fig. 5.13. Adjacent Line voltage space vectors in x1-y1 plane. 142 Fig. 5.14. Adjacent Line voltage space vectors in x2-y2 plane. 143 Fig. 5.15. Three phase source current for 3-7 phase MC. 144 Fig. 6.1. Power circuit topology of 3xN-phase Matrix Converter. 147 Fig. 6.2. Power Circuit topology of three-phase to five-phase Matrix Converter. 154 Fig. 6.3. Input side waveforms of 3 to 5-phase Matrix Converter: (a) Input voltage and
current (b) Spectrum Input current. 156 Fig. 6.4. Output side waveforms of 3 to 5-phase Matrix Converter: (a) Five-phase filtered
output phase voltages (b) Spectrum output voltage. 157 Fig. 6.5. Five-phase series-connected two-motor drive structure. 159 Fig. 6.6. Five-phase parallel-connected two-motor drive structure. 159 Fig. 6.7. Gate signal generation for output phase A. 162 Fig. 6.8. Block diagram of Carrier-based PWM for two frequency output. 162 Fig. 6.9. Input side waveforms of 3 to 5-phase Matrix Converter: upper trace. 163 Fig. 6.10. Input voltage and current, bottom trace, Spectrum Input current. 164 Fig. 6.11. Output filtered five-phase voltages. 164 Fig. 6.12. Spectrum of output voltages; phase ‘A’. 165 Fig. 6.13. Spectrum of output voltages; α-axis voltage. 165 Fig. 6.14. Spectrum of output voltages; x-axis voltage. 166 Fig. 6.15. Response of two-motor drive, a. Speeds, b. torques, c. phase ‘a’ current from
Matrix Converter. 168 Fig. 6.16. Source side voltage and current, voltage is reduced to 150 times. 168 Fig. 6.17i. Matrix Converter output current and voltage time domain and frequency domain
waveform. 169 Fig. 6.18. Power Circuit topology of Three-phase to quasi six-phase Matrix Converter. 170 Fig. 6.19. Stator winding displacement of a quasi or double star six-phase ac machine. 171 Fig. 6.20. Modified offset duty ratios for all input phases. 173 Fig. 6.21. Common mode added reference for output phases. 174 Fig. 6.22. Duty ratio for output phase A. 174 Fig. 6.31. Spectrum of output voltages; a. phase ‘A’, b. α-axis voltage and c. x-axis voltage.
186 Fig. 6.41. Spectrum of output voltages; a. phase ‘A’, b. α-axis voltage c. x1-axis voltage, d.
x2-axis voltage. 200 Fig. 6.43. Output waveforms, filtered output phase ‘a’ voltage. 203 Fig. 7.1. Output and Switching pattern for nth phase in the Case I. 221
xi
Fig. 7.2. Output and Switching pattern for nth phase in the Case II. 222 Fig. 7.3. General Implementation diagram of 3 to k phase Matrix Converter. 223 Fig. 7.4. Output pattern of phase ‘a’. 224 Fig. 7.5. Output pattern of Phase ‘b’. 224 Fig. 7.6. Output pattern of phase ‘c’. 225 Fig. 7.7. Output pattern of phase ‘d’. 225 Fig. 7.8. Output pattern of phase ‘e’. 225 Fig. 7.9. DPWM implementation block without harmonic injection. 227 Fig. 7.10. Input side waveforms of 3 to 5-phase Matrix Converter: a. Input voltage and input
filtered current b. Spectrum input current, c. Input phase current locus. 229 Fig. 7.11. Output side waveforms of 3 to 5-phase Matrix Converter: a. Five-phase output
filtered phase voltages b. Spectrum output filtered voltage, c. locus of α-β axis output voltage, d. locus of x-y axis output voltage, and e. output voltage and current phase ‘a’. 231
Fig. 7.12. Input and output voltage waveforms for a 3 to 5-phase Matrix Converter. 231 Fig. 7.13. Modulation Implementation block using harmonic injection. 233 Fig. 7.14. Input side waveforms of 3 to 5-phase Matrix Converter: a. Input voltage and
current, b. Spectrum input current, c. Input phase current locus. 234 Fig. 7.15. Output waveforms of 3 to 5-phase Matrix Converter: a. Five-phase output filtered
voltages, b. Spectrum output voltage, c. locus of α-β axis output voltage, d. locus of x-y axis output voltage and e. Phase ‘a’ voltage and current. 236
Fig. 7.16. Input and output voltage waveforms for a 3 to 5-phase Matrix Converter with harmonic injection. 236
Fig. 7.17 General Implementation diagram of 3 to 5phase Matrix Converter. 238 Fig. 7.18. General Block diagram of PWM for two frequency output. 239 Fig. 7.19. The input side waveforms; a. source side voltage and current for phase ‘a’, b.
source side three-phase currents, c. converter side three-phase currents. 242 Fig. 7.20. The output side waveform; a. Phase, adjacent and non-adjacent voltages, b. filtered
output phase voltages, c. five-phase inverter currents. 243 Fig. 7.21. The transformed voltage of the output phase voltages: a. and b. Vxy. 244 Fig. 7.22. Time domain and frequency domain waveforms: a. Phase ‘a’ voltage, b. voltage
and c. Vx voltage. 245 Fig. 7.23. Experimental results: tope trace, phase Voltage, middle trace adjacent line Voltage,
and the bottom trace, non-adjacent line Voltage (y-axis: 200V/div, x-axis: 10 msec/div). Error! Bookmark not defined.
xii
LIST OF TABLES
Table 2.1. Reduction in stator copper loss vs phase number of machine. 19 Table 3.1. SWITCHING CONFIGURATIONS USED IN THE SVM ALGORITHM. 39 Table 3.2. Selection of the switching configurations for each combination of output voltage
and input current sectors. 40 Table 4.1. Connectivity matrixfor five-phase two motor drive as shown in reference [4.3]. 56 Table 4.2. Connectivity Matrix. 76 Table 5.1. Large length vectors. 120 Table 5.2. Medium length vectors 122 Table 5.3. Small space voltage vectors. 125 Table 5.4. Three to Six phase output voltage and input current. 130 Table 5.5. The 189 vectors of table.5.4 are distributed as: 134 Table 6.1: Simulation Parameters. 167
xiii
LIST OF ABBREVIATIONS
VSI Voltage Source Inverter
CSI Current Source Inverter
PWM Pulse Width Modulation
MC Matrix Converter
DDPWM Direct Duty Ratio Pulse Width Modulation
DC Direct current
AC Alternating Current
THD Total Harmonic Distortion
DSP Digital Signal Processors
FPGA Field Programmable Gate Arrays
PC Personal Computer
HVDC High Voltage DC
FFT Fast Fourier Transform
RMS Root Mean Square
IGBT Insulated Gate Bi-polar Transistor
LC Inductor Capacitor
WTHD Weighted Total Harmonic Distortion
DMC
Direct Matrix Converter
Mmf Magneto motive force
xiv
LIST OF SYMBOLS
V Voltage Volt (V) I Current Ampere (A) R Resistance Ohms (Ω) α0 Output current space vector - β0 Output voltage space vector - αi Input current space vector - βi Input voltage space vector - ω Angular Velocity rad/sec
aaL Self Inductance between phase a and a Henry (H)
abL Self Inductance between phase a and b Henry (H)
M Mutual inductance Henry (H) Te Electromagnetic torque Newton-meter (Nm)
1
Chapter 1 Introduction
1.1 Preliminary Remark
Variable speed motor drive system is required in numerous industrial applications as they
offer significant advantages compared to a fixed speed drive system. The major advantages
include higher efficiency, better power factor, reduced thermal loading and thus reduced
overall operational costs. Three-phase motor drives are traditionally used in industrial drives
due to readily available three-phase supply and off-the-shelf motor availability. Power
electronic converters are used as interface between three-phase grid supply and the driving
motor. The power converter has no limitation in number of legs being used and thus there is
no restriction in the number of phases of the converters. With the advancement in power
semiconductor technology this additional degree of freedom in power electronic converter is
now exploited by developing multi-phase supply options. Hence multi-phase (more than
three-phase) electric drives have attracted much attention in recent years due to some inherent
advantages that they offer compared to their three-phase counterpart such as lower motor
torque pulsation, less DC link current harmonics, higher redundancy and hence better fault
tolerant characteristics and lower per phase converter ratings etc. The major applications of
multi-phase drives are assumed in safety critical applications such as ship propulsion, ‘more
electric aircraft’ applications, electric and hybrid-electric vehicles, traction, mining, oil & gas
and in general purpose applications in high power range as presented in reference [1.1] -
[1.4].
A block diagram of a variable frequency drive system is shown in Fig. 1.1 The major
components of a variable drive system are: power source (grid supply), power converter for
power processing, controller that can be digital signal processor/micro-controller/field
programmable gate arrays, electric motor, sensors (voltage, current and speed) for feedback
signals. The controller can be programmed using PC.
The power supply to a variable speed drive system is obtained from power electronic
converter that is capable of providing variable voltage and variable frequency. There are four
types of power electronic converters:
i) AC/DC/AC converter with diode based rectifier,
2
ii) AC/DC/AC converter with PWM rectifier also called back-to-back converter,
iii) AC/AC converter called cyclo converter, and
iv) AC/AC converter called Matrix Converter.
This thesis developed control techniques for direct AC/AC converter, i.e. a Matrix
Converter.
Fig. 1.1 General block diagram of a variable speed drive
1.2 Power Electronic Converters
The power electronic converter is the heart of a variable speed drive system. It is used to
process the electrical power of utility grid and supply to the electric motor. This will act as an
interface between the utility grids and the electric motor. Huge research effort is put to
develop technically feasible and commercially viable power electronic converters. The rapid
growth in the semiconductor material and switching devices has led to significant
improvement in the power converters and also has helped in developing their several variants.
Mainly classified, the power electronic converters used in variable speed drive applications
are as presented in reference [1.5];
3
i) AC Voltage Regulator
ii) Cyclo converters
iii) AC-DC-AC converter with diode based rectifier called an inverter
iv) AC-DC-AC converter with active rectifier or PWM rectifier called back-to-back
converter
v) AC-AC converter called Matrix Converter.
AC voltage regulators have limited use since they can only vary the voltage while the output
side frequency is same as the input side frequency. The power semiconductor switching
device used should have bi-directional power flow characteristics. Bi-directional switching
can be obtained by connecting anti-parallel BJTs, or MOSFETs, or IGBTs. In some
applications Triacs and Thyristor based voltage regulators are also used. Other ways of
obtaining bi-directional power flow is shown in Fig. 1.2. This is further elaborated in Chapter
3. The output voltage in AC voltage regulator varies from 0 to input voltage magnitude.
Hence the maximum voltage transfer ratio is 1.
Fig. 1.2. Possible discrete implementations of a bi-directional switch.
Cyclo converter is fully controlled direct AC-AC conversion. Both output voltage magnitude
and frequency are controllable. The maximum output voltage magnitude is same as input
voltage magnitude while the output frequency is limited to 33% of the input frequency.
Hence the application of cyclo converter is also limited but they are used where small speed
control range is needed and is mostly used in high power drive system.
AC-DC-AC converter with diode based rectifier is most commonly employed in variable
speed drive system because of its simplicity and low cost. Several types of diode based
4
rectifiers are in practice but mostly three-phase bridge type rectifiers are the most common.
The output of rectifier contains ripple that can be minimized by using a filter. The output of a
rectifier is used to feed inverter system through DC link capacitor. Large values of DC link
capacitors are usually used to offer constant voltage to the inverter. Rectifiers are
uncontrolled but the inverter is controlled using different types of pulse width modulation.
AC-DC-AC converter with controlled rectifier or active rectifier called back-to-back
converter is also employed where bi-directional power flow is required. The rectifier is
controllable and the power factor can be controlled and can be even made unity. The source
side current is sinusoidal. In case of regeneration of drive system, power can flow back to the
utility grid and this is only possible when active rectifier is used. The output voltage
magnitude is limited by the amount of DC link voltage and the type of PWM method
employed.
Direct AC-AC converter system mostly called Matrix Converter consist of arrays of bi-
directional power semiconductor switches (bi-directional switches are shown in Fig. 1.2).
Three-phase utility grid system is connected to the output through the matrix arrays. Each leg
has three bi-directional switches and any output can be connected to any input line through
the switching action of bi-directional power switches. The voltage of input side appears at the
output side and the current in any phase of the load can be drawn from any phase of the
utility grid. A small LC filter is connected at the source side to remove the current ripple
(Which appears due to switching action). Matrix Converter has the following major
advantages:
Sinusoidal input and output currents
Controllable input side power factor
No bulky DC link capacitor is needed
Bi-directional power flow.
The major disadvantage with the Matrix Converter is the low output voltage, in case of three-
phase configuration the output is 15% less than the input side. In case of five-phase output
the output is almost 20% lower than the input side. Higher the output number of phases,
lower the output voltage magnitude. To enhance the output voltage, over-modulation is
required and also AC chopper can be used in conjunction with the Matrix Converter.
Additional shortcoming of a Matrix Converter is its complex control.
5
1.3 Features of Multi-phase Motor Drive
An obvious question arises, why the thesis topic is chosen on multi-phase motor drive when
three-phase drive is being successfully used in industries for decades? The simple answer lies
in the inherent advantages that a multi-phase drive offers compared to their three-phase
counterpart. Some of the known advantages of multi-phase motor drives are given in
reference [1.1]:
a. Reducing the amplitude of torque pulsation and increasing the frequency of torque
pulsation in inverter fed multi-phase drive system when inverter is operating in square
wave mode. The frequency of torque pulsation is 2n*fundamental frequency, where n
is the number of phases. Thus for instance in a five-phase machine the torque
pulsation occurs at 10 times the fundamental frequency while in three-phase case it is
6 times the fundamental.
b. .Higher efficiency compared to the three-phase counterpart. This is attributed to the
fact that the stator excitation produces a field with lower space harmonic in case of
multi-phase machine when compared to three-phase machines.
c. Higher torque density in multi-phase machines compared to the three-phase machines.
The reason behind this is that apart from fundamental, higher current harmonic
contribute towards the torque development in concentrated winding machines. For
instance in case of a five-phase machine, third harmonic along with the fundamental
may be injected to enhance the torque production and similarly in case of a seven-
phase machine, 3rd and 5th harmonics may be utilised. Thus in general harmonics
lower than the phase number may be utilised effectively to enhance the torque
production. This is a special characteristic of multi-phase machines and is not
available in a three-phase machine.
d. Greater fault tolerance than their three-phase counterpart and offers more reliable
solution. If one phase of a three-phase opens, the machine may continue to run but it
requires special arrangement for starting (i.e. a divided DC bus with a neutral point)
and the machine has to be heavily de-rated to avoid excessive heating. In contrast in
case of five-phase machine, the machine will start, accelerated, reject load transient
and continue to run normally with minimal de-rating even with a loss of one phase. In
6
case of a seven-phase machine unnoticeable change occurs even with a loss of up to
two phases. This trend continues with higher phase number. Thus multi-phase
machine drive is suited ideally for safety critical applications such as ship propulsion,
air craft applications & defence and emergency services applications.
e. Less volume to weight ratio
f. Better noise and vibration characteristics
g. Lower DC link current harmonics
h. Reducing rotor current harmonics
i. Reducing current per-phase without increasing voltage per-phase. This reduction in
power per-phase translates into the reduction in the rating per converter leg. Thus the
series/parallel combination of power semiconductor switches may be avoided and
consequently avoiding the associated static and dynamic voltage sharing problems.
This is one of the driving forces behind the accelerated use of multi-phase machines
in high power drive applications.
1.4 Analyzed Drive Topologies
In this thesis, multi-phase multi-motor drive system is analyzed. The focus of the research is
on the development of control schemes for multi-phase Matrix Converter feeding multi-phase
multi-motor drive system. When referring to multi-motor drive system, means more than one
machine are controlled simultaneously either independently or under identical operational
conditions. In three-phase multi-motor drive system, two topologies are possible:
i) X number of three-phase motors with independent vector control, which is fed using
X number of three-phase inverters with 3X legs. Motor/inverter sets connected in
parallel, with common DC link. This configuration is shown in Fig. 1.3a.
ii) X number of three-phase motors which is fed using one three-phase inverter. In this
structure motors have to be identical and to operate under the same load torque with
the same angular speed. This is shown in Fig. 1.3b.
Both these drive topologies do not offer independent control of connected three-phase
motors.
7
Vector Feedback control common DC link 3-phase PWM Machine VSI 1 3-phase Machine PWM 2 VSI Vector control Feedback
a b
Fig. 1.3. Three-phase multi-motor drive system.
Multi-phase drive system can however, offer independent control of more than one machine
when supplied from a single variable voltage and variable frequency power source and field
oriented control is employed as shown in reference [1.6].
This is a generalized concept where multi-phase machine’s stator can be either connected in
series or in parallel while supply is given from only one power source and all the machines
can be controlled independently both in transient and steady state conditions. The topologies
investigated in this thesis are:
Five-phase series-connected two-motor drive system
Six-phase series-connected two-motor drive system
Seven-phase series-connected three-motor drive system.
It is important to mention here that the main purpose of the thesis is to develop pulse width
modulation techniques for three-phase input and multi-phase (5-phase, 6-phase, 7-phase)
output to produce appropriate fundamental frequencies. Further to note, that the developed
PWM are equally applicable to series-connected drive and parallel-connected drive systems.
Series-connected or parallel-connected refers to stator winding connection and rotors are
independent.
It is evident from the dynamic model of multi-phase machine in given reference [1.7], that
only two orthogonal current components namely d-q is responsible for torque production and
rest of the components do not contribute to the torque production rather they are loss
VectorControlVectorControl
3-phasePWMVSI
Machine1
Machine2
Feedbacksignals
Feedback signals
DC Link
8
producing and they are limited only by leakage inductances. This finding is important as the
field oriented control principle of a three-phase machine can be extended very easily to the
multi-phase machine as shown in reference [1.6]. The extra non-torque current components
are further utilized recently to control other machine in series/parallel-connected multi-motor
drive system presented in references [1.8] - [1.12]. This is a general concept where n number
of multi-phase machine’s stator windings can be connected either in series or parallel and the
supply is given from a single multi-phase voltage source inverter and all the machines in the
group can be controlled independently using field oriented control principle. The number of
connectable machines depends upon the phase number and if the phase number is even or
odd for instance in a five-phase and six-phase system two machines can be connected, in
seven and eight-phase system three machines can be connected and controlled independently.
The specific case of five-phase system is presented in reference [1.10-1.11], six-phase system
is presented in reference [1.8-1.9] and seven-phase system is presented in reference [1.12].
The basic principle of control lies in the fact that the d-q component (torque producing) one
machine becomes the x-y (non-torque producing) component of the other and vice versa. This
is possible by appropriate phase transposition in the connection of the two or more stator
windings. The major advantage of such concept is reduction in the number of power
converters used, e.g. in case of a five-phase machine, instead of using two separate converters
just one converter is sufficient. The major disadvantage of this concept is increase in the
losses because current of all machines is carried by all the stator winding of the machines.
This is worst in case of five- and seven-phase systems but it is marginally better in case of a
six-phase system where a six-phase machine and three-phase is connected in series. The six-
phase machine’s stator current cancels out at the point of interconnection of three-phase
machine. Thus the losses of six-phase machine will increase marginally if small three-phase
motors are used and three-phase motor’s losses will not alter. The application of a five-phase
two-motor drive system is identified in a winder drive where the winding and unwinding
machines can be the two series-connected machines. In this special application, situation
never arises when both the machines are loaded to rated condition. When load on one
machine increases, the load on the other machine decreases. The application of six-phase
two-motor drive system is identified in a situation where six-phase machine can be used as
main driving motor such as in ship propulsion and three-phase machine can be used for
auxiliary function such as for pumping. The basic structure of five-phase series-connected
two motor drives, six-phase series-connected two motor drives and seven-phase three-motor
9
drive are shown in Fig. 1.4. These three drive system topologies are used in the present thesis
for developing their control using direct AC-AC multi-phase Matrix Converter.
AI
BI
CI
DI
EI
1 2
Fig. 1.4a. Five-phase series-connected two-motor drive structure.
AI
BI
CI
DI
EI
1 2
FI
Fig. 1.4b. Six-phase series-connected two-motor drive system.
AI
BI
CI
DI
EI
1
FI
2ai
2bi
2ci
2di
2ei
2
2fi
3ai
3bi
3ci
3di
3ei
3
3fi
3gi2giGI
Fig. 1.4c. Seven-phase series-connected three-motor drive system.
10
Since the thesis explores the Pulse Width Modulation (PWM) of AC-AC Matrix Converter
for controlling multi-phase multi-motor drive system, it is important to present state-of-the-
art in the PWM techniques of such converter. A review on the state-of-the-art in the Matrix
Converter is presented in references [1.13] - [1.16]. Power circuit of a general three-phase to
n-phase Matrix Converter is shown in Fig. 1.5. The determination of appropriate PWM
method for complex Matrix Converter topology especially for multi-phase output is a key
issue. The restriction imposed on the PWM strategies are: the input phases must not be short
circuited and the output phase should not be open circuited.
Fig. 1.5. Outlay of a three-phase to n-phase direct Matrix Converter.
The early modulation method proposed by Alesina and Venturuni [1.17] - [1.18] for three-to
three-phase Matrix Converter may still be extended for multi-phase Matrix Converter
topology with some modifications, but the output voltage magnitude will once again be
limited to 50% of the input voltage magnitude. Thus it is not practical to use this method due
to small output voltage magnitude. Similar concept is put forth in reference [1.19], where the
outputs are dealt independently and PWM method is presented focusing on three-to single-
phase and three-to three-phase Matrix Converter. This thesis describes the application of
optimum control theory to N-input K-output Matrix Converters based on the transformation
of actual converter topology into a suitable equivalent structure. The topologies and control
properties of the most common Matrix Converters are analyzed by developing the general
anti-transformation criteria.
Space vector PWM approach used for three-to three-phase Matrix Converter, offering higher
output voltage magnitude as references [1.20]-[1.21], although can be extended to multi-
phase topology as presented in references [1.22]-[1.24], is highly complex due to large
11
number of space vectors available. For instance in a three-to five-phase Matrix Converter
total number of space vector generated are 215 = 32768, nevertheless, due to constraints
imposed by switching actions, 35 =243 vectors are useful. Following the analogy with five-
phase three-level voltage source inverter, the number of space vectors that can be used to
generated sinusoidal output is further limited to 93 as in reference [1.22]. Nevertheless, space
vectors PWM for real time processing still poses a challenge. The output voltage magnitude
with space vector PWM in a three to five-phase Matrix Converter is limited to 78.86% of the
input magnitude for linear modulation range. This limit can further be raised by exploiting
over-modulation region. No work has so far been reported on over-modulation of the multi-
phase Matrix Converter and also this is not the subject of this thesis.
Carrier-based PWM scheme is the simplest approach for controlling a Matrix Converter.
Carrier based PWM is one of the recently developed modulation strategies for the Matrix
Converter in three-to three-phase topology. As in the traditional sinusoidal PWM used in
voltage source inverters, it employs the carrier and reference signals. The implementation of
this scheme is simpler compared to the space vector PWM by using conventional UP/DOWN
counters, which are embedded in most of the single chip microcontrollers. However, the
carrier signal employed in this PWM is discontinuous and thus the method is intuitively
difficult to understand. In addition, increasing the limit of modulation is a trivial task. Since,
offset is added in this PWM, it is not suitable for topology where input and output neutral
needs connection.
Another simpler and modular modulation approach, compared to the space vector PWM was
developed called ‘direct duty ratio PWM’, which offers the advantages of simpler real time
implementation as shown in reference [1.25]. This approach is modular in nature and it is
seen that this can be employed for any phase number or switch number of the Matrix
Converter. The major advantage of this modulation scheme is that it is highly intuitive and
flexible, and can be applied to any topology of the Matrix Converter as the concept is based
on ‘per output phase’.
This thesis focuses on direct AC-AC three-phase to n-phase Matrix Converter fed
series/parallel-connected multi-motor drive system. The emphasis is placed on the
development of appropriate PWM techniques for five-phase, six-phase two-motor drive and
seven-phase three-motor drive (Fig. 1.4). The major aim is to develop two fundamental
frequency output for two-motor drive and three fundamental frequencies output for three-
motor drive from multi-phase Matrix Converter. The pulse width modulation techniques
developed in this thesis are:
12
Carrier-based PWM
Direct duty ratio based PWM.
1.5 Research Objectives
The present research is based on the development of mathematical model of three-phase input
and multi-phase output direct AC-AC Matrix Converter and exploring their control algorithm
for the multi motor drive system. The principle set objectives of the proposed research are:
i) To investigate the current state-of-the art in multi-phase motor drive research area by
carrying out comprehensive literature review from available data bases.
ii) To review the modelling procedure of a five-phase series-connected two-motor drive,
a six-phase and three-phase series-connected two-motor drive and a seven-phase
three-motor drive using general machine modelling approach.
iii) To review the modelling procedure of a three-phase to five-phase Matrix Converter, a
three-phase to six-phase Matrix Converter and three-phase to seven-phase Matrix
Converter using space vector concept.
iv) To investigate the operation of a three-phase input and five-phase output Matrix
Converter for producing two independent fundamental frequency components using;
i) carrier-based PWM, ii) direct duty ratio based PWM assuming series-
connected/parallel-connected two motor drive system being fed using this power
source.
v) To investigate the operation of a three-phase input and six-phase output Matrix
Converter for producing two independent fundamental frequency components using
carrier-based PWM, assuming series-connected/parallel-connected two motor drive
system being fed using this power source.
vi) To investigate the operation of a three-phase input and seven-phase output Matrix
Converter for producing three independent fundamental frequency components using
carrier-based PWM, assuming series-connected/parallel-connected three motor drive
system being fed using this power source.
1.6 Layout of the Thesis
The complete thesis is organized into eight different chapters. Chapter 1 starts with the
introductory remark and provides a review of basic literature available on the specific topic of
13
multi-phase motor drives. An overview on the existing power electronic converter topologies
is presented. A question is raised as to why at all the proposed research is looking into the
multi-phase motor drives and in support a number of existing advantages are highlighted. An
overview of the drive configuration investigated in the thesis is elaborated. Brief introduction
to the Matrix Converter is elaborated. The objectives of the research are set and are
discussed.
Chapter 2 presents literature review in the multi-phase multi-motor drive control research. At
first state-of-the art in multi-phase drive system is discussed. This is followed by the
description on conventional Matrix Converter topology and its control is elaborated. Finally
multi-phase Matrix Converter topology is discussed along with its control and work done so
far in the literature.
Chapter 3 is dedicated to the existing control technique of conventional three-phase to three-
phase direct Matrix Converter. The control can be divided into scalar and vector control
approaches. The limitation of lower output voltage is rectified by harmonic injection.
Theoretical maximum limit of output voltage is achieved and elaborated.
Chapter 4 is dedicated to the modelling of a multi-phase series-connected multi-motor drive
system. Three configurations are taken up for discussion: five-phase series-connected two-
motor, six-phase two-motor and seven-phase three-motor. Modelling is done assuming multi-
phase induction machines. Although this concept of series connection is independent of type
of machines, but only one type of machine is considered to show the viability of the
approach. Phase variable modelling approach is used with general assumption of sinusoidally
distributed mmf. This is followed by transformation of the model in the rotational reference
frame and then in the stationary reference frame. The developed model is validated using
MATLAB/SIMULINK approach assuming ideal sinusoidal supply.
Chapter 5 elaborate the space vector model of multi-phase Matrix Converter. Three different
topologies are discussed: three-phase to five-phase, three-phase to six-phase and three-phase
to seven-phase. For n-phase input and m-phase output, the number of switching states is 2nxm
and after imposing the constraints, the remaining switching states are elaborated in this
chapter.
The PWM approaches of multi-phase Matrix Converter are complex and challenging due to
their circuit configuration and AC signals at both input side and output side. Hence, simpler
14
approach of carrier-based PWM is discussed in Chapter 6. At first carrier-based scheme for
single-motor drive is elaborated followed by multi-phase multi-motor drive system. Five-
phase single-and two-motor drive, six-phase single and two-motor drive and seven-phase
single-and three-motor drive topologies are discussed. Simulation and experimental results
are included to validate the PWM schemes.
Chapter 7 elaborate another PWM technique based on direct duty ratio calculation. This is a
simple approach that is highly modular in nature since the computation is done one per-leg
basis. Five-phase single-motor and two-motor drive system is discussed. Simulation and
experimental results are included.
Chapter 8 is the final chapter, and provides a summary of the thesis and the salient points
from each chapter. Conclusions are made as to the viability of the series-connected/parallel-
connected multi-phase multi-motor drive fed using three-phase input multi-phase output
direct Matrix Converter. Future research work possible related to this area is suggested and
reported in this chapter.
References used in each chapter are given at the end of each chapter itself and hence no
separate references chapter is given. List of publications out of this thesis is provided at the
end.
1.7 Novelty and Contribution of Thesis
The major contributions of the thesis are as follows:
Development of mathematical model of multi-phase multi-motor drive system. Mathematical model of a five-phase two-motor drive and a six-phase two-motor drive is obtained from the literature and is further developed in and reported in the thesis. Mathematical model of seven-phase three-motor drive system is not reported in the literature and is developed in this thesis based on phase variable concept. The developed model is then transformed into three orthogonal planes using general transformation matrix. The mathematical model developed is verified using simulation approach. The developed mathematical model helps in developing decoupled control scheme.
Development of simple control algorithms for decoupled dynamic control of five-phase two-motor drive, six-phase two-motor drive and seven-phase three-motor drive. Control algorithm is developed based on sinusoidal carrier-based PWM scheme and direct duty ratio based PWM scheme.
15
References:
[1.1] E. Levi, “Multi-phase Machines for Variable speed applications” IEEE Trans. Ind. Elect., vol. 55, no. 5, May 2008, pp. 1893-1909.
[1.2] G.K. Singh “Multi-phase induction machine drive research – a survey”, Electric Power System Research, vol. 61, 2002, pp. 139-147.
[1.3] R. Bojoi, F. Farina, F. Profumo, Tenconi,“Dual three induction machine drives control-A survey”, IEEE Tran. On Ind. Appl.,vol. 126, no. 4, pp. 420-429, 2006.
[1.4] E. Levi, R. Bojoi, F. Profumo, H.A. Toliyat, S. Williamson, “Multi-phase induction motor drives-A technology status review”, IET Elect. Power Appl. Vol. 1, no. 4, pp. 489-516, July 2007.
[1.5] H. Abu-Rub, A. Iqbal, J. Guzinski, “High Performance Control of AC Drives with MATLAB/SIMULINK Models”, Wiley, UK, 2012.
[1.6] A. Iqbal, “Modelling and control of series-connected five-phase and six-phase two-motor drives,” PhD Thesis, Liverpool John Moores University, Liverpool, UK, 2006.
[1.7] E. Levi, M .Jones, S.N. Vukosavic, H.A. Toliyat, “A novel concept of a multi-phase, multi-motor vector controlled drive system supplied from a single voltage source inverter,” IEEE Trans. on Power Electronics, vol. 19, no. 2, 2004, pp. 320-335.
[1.8] E. Levi, S.N. Vukosavic, M. Jones, “Vector control schemes for series-connected six-phase two-motor drive systems,” IEE Proc. – Electric Power Applications, vol. 152, no. 2, 2005, pp. 226-238.
[1.9] M. Jones, S.N. Vukosavic, E.Levi, A.Iqbal, “A six-phase series-connected two-motor drive with decoupled dynamic control,” IEEE Trans. on Industry Applications, vol. 41, no. 4, 2005, pp. 1056-1066.
[1.10] E. Levi, M. Jones, A. Iqbal, S.N. Vukosavic, H.A. Toliyat, “An induction machine / Syn-Rel two-motor five-phase series-connected drive,” IEEE Trans. on Energy Conversion, vol. 22, no. 2, 2007, pp. 281-289.
[1.11] E. Levi, M. Jones, S.N. Vukosavic, A. Iqbal, H.A. Toliyat, “Modelling, control and experimental investigation of a five-phase series-connected two-motor drive with single inverter supply,” IEEE Trans. on Industrial Electronics, vol. 54, no. 3, 2007, pp. 1504-1516.
[1.12] M. Jone, E. Levi, S.N. Vukosavic, H.A. Toliyat, “Independent vector control of a seven-phase three-motor drive system supplied from a single voltage source inverter”, Proc. 34th IEEE PESC, 2003, vol. 4, pp. 1865-1870.
[1.13] P.W. Wheeler, J. Rodriguez, Jon C. Clare, L. Empringham, A. Weinstein, “Matrix Converters: A Technology Review”, IEEE Trans. On Ind. Elect. vol. 49, no. 2, April, 2002, pp. 276-288.
[1.14] P. Wheeler, L. Xu, M. Y.Lee, L. Empringham, C. Klumpner, J.Clare, “A review of multi-level Matrix Converter topologies,” Proc. IET Power Electronics, Machines and Drives Conf., York, UK, 2008, pp. 286-290.
[1.15] L. Empringham, J. Kolar, J. Rodriguez, P. W. Wheeler, J. C. Clare, "Technological issues and industrial application of Matrix Converters: A review," IEEE Trans. on Ind. Elec., 2013 (d.o.i.10.1109/TIE.2012.2216231).
[1.16] J. Rodriguez, M. Rivera, J. W. Kolar, P. W. Wheeler, "A review of control and modulation methods for Matrix Converters," IEEE Trans. On Ind. Elec., vol. 59, no. 1, pp. 58-70, 2012.
[1.17] A. Alesina, M. Venturini, “Solid state power conversion: A Fourier analysis approach to generalised transformer synthesis” IEEE Trans. Circuit System, vol. 28, no. CS-4, pp. 319-330, April 1981.
16
[1.18] A. Alesina, M. Venturini, “Analysis and design of optimum amplitude nine-switch direct AC-AC converters”, IEEE Trans. Power Elect. vol. PE-4, no. 1, pp. 101-112, 1989.
[1.19] P. Tenti, L. Malesani, L. Rossetto, “Optimum control of N-input K-output Matrix Converters,” IEEE Trans. on Power Electronics, vol. 7, no. 4, 1992, pp. 707-713.
[1.20] D. Casadei, G. Grandi, G. Serra, A. Tani, “Space vector control of Matrix Converters with unity power factor and sinusoidal input/output waveforms”, Proc. EPE Conf., vol. 7, pp. 170-175, 1993.
[1.21] L. Huber, D. Borojevic, “Space vector modulated three-phase to three-phase Matrix Converter with input power factor correction”, IEEE Trans. Ind. Appl. Vol. 31, no. 6, pp. 1234-1246, Nov./Dec. 1995.
[1.22] Sk. M. Ahmed, A. Iqbal, H. Abu-Rub, M.R. Khan, “Space vector PWM technique for a novel three-to-five phase Matrix Converter,” IEEE Energy Conversion Conf. and Exhibition, Atlanta, GA, 2010, pp. 1875-1880.
[1.23] Sk. M. Ahmed, A. Iqbal, H. Abu-rub, M.R. Khan, “Carrier based PWM technique for a novel three-to-seven phase Matrix Converter,” Int. Conf. on Electrical Machines ICEM, Rome, Italy, 2010, CD-ROM.\
[1.24] A. Iqbal, Sk.M. Ahmed, H. Abu-Rub, M.R. Khan, “Carrier based PWM technique for a novel three-to-five phase Matrix Converter,” Power Conversion and Intelligent Motion PCIM, Nuremberg, Germany, 2010, pp. 998-1003.
[1.25] Y. Li, N.-S. Choi, B.-M. Han, K. M. Kim, B. Lee, J.-H. Park, "Direct duty ratio pulse width modulation method for Matrix Converters, "Int. Journal of Control, Automation, and Systems, vol. 6, no. 5, pp.660-669, 2008.
17
Chapter 2 Literature Review
2.1 Introduction
Traditionally, variable-speed electric drives are based on three-phase configuration of an
electric machine and a power electronic converter. Such a situation is the relic of a bygone
era, when machines were supplied from the grid. Since the power electronic converter can be
viewed as an interface that decouples three-phase mains from the machine, the number of
machine’s phases does not have to be limited to three in variable-speed drives as shown
reference [2.1]. Nevertheless, three-phase machines are still customarily adopted for variable
speed applications due to the wide off-the-shelf availability of both machines and converters.
However, in all applications where a machine is not readily available, multi-phase machines
(machines with more than three phases on stator) offer a number of advantages as shown in
reference [2.2]. These include (but are not limited to): i) possibility of splitting the required
per-phase power rating across more than three phases, thus reducing current rating of the
semiconductor components (of exceptional importance in high-power and high-current
applications), ii) a significant improvement in fault tolerance of the drive, since any AC
machines (regardless of the number of phases) requires only two currents for independent
flux and torque control of the machine (in a three-phase machine open-circuit fault in one
phase means that there are not two independent currents available for control anymore;
however, an n-phase machine can continue to operate with a rotating field in post-fault
operation as long as no more than (n-3) phases are faulted), and, iii) a potentially better
efficiency due to reduced space harmonic content of the magneto-motive force.
Although the multi-phase variable-speed drive systems have been a subject of research
interest for the last 50 years, it is the last ten years that have seen an enormous growth of the
quantum of knowledge in the area as presented in reference [2.1]. This has been initiated by
numerous specific application areas, such as high power industrial applications, electric ship
propulsion, locomotive traction, electric and hybrid electric vehicles, etc., where advantages
of multi-phase systems outweigh the initial higher cost of the development.
In general, the conversion of an input ac power at a given frequency to an output power at a
different frequency can be obtained with different systems, employing rotating electrical
machinery, non-linear magnetic devices or static circuits containing controllable power
18
electronic switches. Restricting the discussion of AC/AC power frequency conversion to
static circuits, the available structures can be divided in “direct” and “indirect” power
conversion schemes. Indirect schemes consist of two or more stages of power conversion and
an intermediate DC-link stage is always present. A typical example of two stage indirect
AC/DC/AC power frequency conversion is the diode-bridge rectifier-inverter structure, in
which an AC power is firstly converted to a DC power (diode rectifier), and then converted
back to an AC power at variable frequency (inverter). In direct conversion schemes a single
stage carries out the AC/AC power frequency conversion. This is the subject of the research.
2.2 Multi-phase Motor Drive Systems
Multi-phase ac machine can be either induction or synchronous. The induction machine can
be either a wound or cage type, it is the later design which has been extensively discussed in
the literature. In synchronous category it can either be a permanent magnet, with field
winding or reluctance type. The stator of a multi-phase can be designed with sinusoidal
winding or with concentrated winding. Both machines differ in properties due to different
distribution of their mmf waves. Some of the properties of multi-phase machines which are
independent of their type and are highlighted in references [2.3]-[2.10].
MMF produced by stator excitation have lower space harmonics and this affect a number of
characteristics of a multi-phase machine such as their efficiency and power factor.
The frequency of torque ripple in a multi-phase machine fed using a multi-phase
inverter operating in square wave mode is 2n, where n being phase number, thus
increasing phase number correspondingly increases the frequency of torque ripple.
Only two current components are required for the torque/flux control of an ac
machine irrespective of number of phases thus the extra current components available
in multi-phase machines can be utilised for other purposes.
Due to sharing of power among large number of phases the power per phase gets
reduced for handling same amount of power as that of a three-phase machine. The
power sharing by each inverter leg also reduces and thus lower rating power
semiconductor can be used.
19
Due to availability of large number of phases a multi-phase machine is more fault
tolerant compared to their three-phase counterpart. Since only two current
components are required for torque/flux control a multi-phase machine do not pose
any problem in the event of occurrence of fault unless there is a loss of more than n-3
phases occurs.
The consequence of mmf with lower space harmonic in multi-phase machine is their higher
efficiency and lower acoustic noise. Higher frequency of torque ripple once again put less
stress on the driving load and quieter operation of the machine. Higher efficiency of multi-
phase machine compared to three-phase machine may be explained as follows: consider two
machines of identical design with different phase numbers. If both machines have to develop
same amount of torque at same speed, then they must have same rotor copper losses, same
air-gap mmf and same fundamental component of stator current. The higher efficiency is
attributed to lower stator copper losses. The stator copper losses will reduce with increasing
phase number as given in Table [2.1].
Table 2.1. Reduction in stator copper loss vs phase number of machine.
Phase Number 5 6 9 12 15
Reduction in stator Cu. Loss 5.6% 6.7% 7.9% 8.3% 8.5%
It is important to note here that the rotor copper losses and iron losses (since air gap mmf is
unchanged) will remain unchanged with the change in machine phase number. The reduction
in stator copper losses is due to that fact that the stator current lower order harmonic changes.
The use of higher phase number increases the pole number of the harmonic components and
thus reducing their magnitude and consequently reducing the stator copper losses due to these
harmonic components.
Reduction in the magnitude of torque pulsation and the increase in the frequency of torque
pulsation in an inverter (operating in 180° conduction mode) fed multi-phase is their another
salient feature. as in reference [2.11], it is demonstrated that the air gap field produced by qth
harmonic component of excitation current in a 2P pole machine with n balanced phases will
have pole-pair numbers given by:
( 2 ) 0, 1, 2, 3,......P q kn k (2.1)
Positive values of β correspond to forward rotating fields and negative values represents
backward rotating field. For fields of the same pole number to be produced by two distinct
20
excitation components , with frequencies f1 and f2, (2.1) shows that there must be solution to
the equation:
nkqPnkqP 2211 22 (2.2)
from which one gets:
212121 222 kknnknkqq (2.3)
Thus the frequency of torque pulsation is:
2121 2 kknffqq (2.4)
where k1 and k2 are integers. Equation (2.4) reveals that the frequency of torque pulsation in a
balanced n-phase machine is produced at all even multiples of the product of the phase
number and the fundamental frequency of the supply. For instance in a three-phase machine
the torque ripple frequency exist at multiple of 300 Hz, in a five-phase machine it shifts to
500 Hz . Thus higher the phase number higher will be the torque pulsation frequency.
One important property of a multi-phase machine with concentrated stator winding is
enhancement in average torque produced by the machine by injecting higher order harmonic
currents [2.12-2.15]. This is a special characteristic of a multi-phase machine which is not
available in a three-phase machine. All harmonic current of the order between fundamental
and n can be injected along with fundamental to improve the torque production. For instance
in a five-phase machine 3rd harmonic can be added and in a seven-phase machine 3rd and 5th
can be added with the fundamental as presented in reference [2.1]. In even phase number
machine only quasi six-phase configuration is utilised for torque enhancement by injecting 3rd
harmonic as shown in reference [2.16].
Assuming an odd phase number machine with one neutral point, there exist n-3 additional
current components and thus the same number of additional degree of freedom. This
additional degree of freedom may be used for various purposes, such as:
Enhancement of torque production by injecting higher order current harmonics
in concentrated winding machines as above mentioned.
Fault mitigation and providing fault tolerant operation. A number of
publications are available in this regard, some of them are given in reference
[2.17-2.27]. In the event of a fault (one phase out) in a three-phase drive with
21
star-connected stator winding and isolated neutral point, the two available
healthy phase currents become identical with 180° phase shift. Hence it is no
more possible to independently control the remaining two phase currents unless
a divided DC bus along with a neutral point is provided as demonstrated in
reference [2.28]. In contrast to this, multi-phase machine may still generate
rotating mmf with the loss of upto n-3 phases. The phase redundancy concept
was developed by Jahns in reference [2.29] for multi-phase machine and it was
shown that machine normal operation of a multi-phase machine is possible with
appropriate post fault strategy. The simplest post fault strategy can be adopted
for a machine with n=pk phases (p = 3,4,5…., k = 2,3,4,…) and the complete
winding is configured as k windings with a phase each, with k isolated neutral
points and k independent p-phase inverters. In the event of a fault in any phase
of p set, then the complete set of p-phases is taken out of the service. For
instance in case of quasi six-phase machine with two set of three-phase
windings one set is completely taken out of the service rendering the supply to
only one set and thus the complete drive operate with half torque and power. In
case of a 15-phase motor with three set of 5-phase windings, one or more
complete 5-phase windings may be taken out of the service to keep the drive
running with reduced rating. This simple strategy is although easy to implement
and no change is required in the software but this may not be suitable for safety
critical applications as there is a huge loss in developed torque. In such
situations, machine with one neutral point is better suited. This is so since the
currents of all the remaining healthy phases are regulated to offer an optimum
solution. The post fault strategy is application dependent and three different
situations may occur as highlighted in reference [2.25]. Assuming loss of one
phase, strategy 1 could be to maintain the same torque level as that of the pre
fault conditions without any pulsation. In this case the increase in currents in
healthy phases are inevitable and the currents increase by a factor 1n
n and
correspondingly the stator copper losses will also increase but there will be no
change in the rotor copper losses. The second strategy is based on the keeping
the stator copper loss under post-fault condition to the same level as that of the
pre-fault condition. The stator current has to be increased to 1n
nin the
22
remaining healthy phases. The torque and power reduces and the rotor copper
loss increases. The third strategy keeps the stator current during post fault at the
same level as that of pre fault condition. In this scheme the stator copper losses
reduced by a factor of 1n
n , however, the torque level and power is reduced
with corresponding increase in the rotor copper loss. All three scheme discussed
in reference [2.25] assumes load whose torque varies as square of speed. All
these strategies require configuring the software in post-fault operation.
Another useful purpose of additional degree of freedom in terms of extra current
components is the independent control of two or more multi-phase machine
whose stator windings are connected in series or parallel. A number of
published work is available on this issue which is taken up in detail in next
section
It is important to emphasize here that the additional degree of freedom available in multi-
phase motor drive can be utilised for one purpose at one time as given in reference [2.1].
2.3 Three-phase input and three-phase output Matrix Converter
A direct AC-AC converter is proposed in reference [2.30] called ‘Matrix Converter’ has been
the focus of an intensive on-going research. Several aspects of the Matrix Converter has been
investigated in the literature. Most commonly the control techniques have been explored, and
comprehensive review is presented in references [2.31-2.32]. From the literature review, it is
seen that the control characteristics of a direct AC-AC Matrix Converter when used in
variable speed drives application, are:
The output voltage magnitude and frequency should be controlled independently;
The source side (utility grid side) current should be completely sinusoidal and the
power factor angle of the source side should be fully controllable
The output voltage to input voltage ratio should be maximum
The Matrix Converter should satisfy the conflicting requirements of minimum low
order harmonics distortion and minimum switching losses
The Matrix Converter input side is connected to the utility grid and hence prone to
unbalancing and distortion. The control should compensate for such operational
conditions
The control should be computationally efficient.
23
The control scheme originally proposed in reference [2.33], although characterized by better
performance than naturally commutated cyclo-converters, it had significant output voltage
limitations and serious output voltage waveforms distortion with high total harmonic
distortion.
An alternative high switching frequency control technique was proposed in reference [2.34]
which was more effective than traditional control techniques, but still can generate a
maximum output voltage equal to half of the input voltage magnitude. This is a scalar control
approach and had serious limitation of reduced output voltage and practically had no
attraction. Further to the this, other limitation is the poor factor control at the source side.
Another control approach was proposed in references [2.35-2.37], where a Matrix Converter
is considered as composed of two stages i.e. rectification and inversion. Hence the input
voltages are first “rectified” to create a fictitious DC bus and then “inverted” to obtain
variable voltage and variable frequency waveform. However, the limitation of output voltage
magnitude and uncontrolled source side power factor was still un resolved.
The direct PWM method developed by Alesina and Venturini [2.34] limits the output to half
the input voltage. This limit was subsequently raised to 0.866 by taking advantage of third
harmonic injection as shown in reference [2.38] and it was realized that this is maximum
output that can be obtained from a three-to-three phase Matrix Converter. Indirect method
assumes a Matrix Converter as a cascaded virtual three-phase rectifier and a virtual voltage
source inverter with imaginary DC link. With this representation, space vector PWM method
of VSI is extended to a Matrix Converter as given in references [2.39]-[2.40]. Although the
space vector PWM method is suited to three-phase system but the complexity of
implementation increases with the increase in the number of switches/phases. Motivated from
the simple implementation, carrier-based PWM scheme has been introduced for Matrix
Converter as shown in reference [2.41-2.43]. Carrier based PWM is the latest modulation
strategy for the Matrix Converter. As in the traditional sinusoidal PWM used in voltage
source inverters, it employs the carrier and reference signals. The implementation of this
scheme is simpler using conventional UP/DOWN counter which are embedded in a single
chip microcontrollers. This chapter focuses on different Matrix Converter modulation
strategies based on space vector modulation.
2.4 Multi-phase Matrix Converter
The power circuit topology of a three-phase to k-phase Matrix Converter is illustrated in Fig.
2.1. There are k legs with each leg having three bi-directional power switches connected in
24
series. Each power switch is bi-directional in nature with anti-parallel connected Insulated
Gate Bi-polar Transistors (IGBTs) and diodes. The input is similar to a three-phase to three-
phase Matrix Converter having LC filters.
AC
AC
AC
Isb
VAVB VC
Va
Vb
VC
Ia
Ib
Ic
Input Filter
Isa
Isc
S11 S12 S13
S21 S22 S23 S2k
S1k
S31 S32 S33 S3k
Input
K phase Output
Vk
L
L
L
C C C
Fig. 2.1. Power Circuit topology of Three-phase to k-phase Matrix Converter.
The load to the Matrix Converter could be a star-connected R or R-L or an AC machine. As
far as the research on multi-phase output AC-AC converters is concerned, there has been
relatively little development until recently. Theoretical principles of such a converter have
been established in reference [2.44]. Probably the first considerations of a real world
application of a three to n-phase Matrix Converter application, in more-electric aircraft, have
been reported as shown in reference [2.45]. During the last four years or so, Matrix Converter
25
with multi-phase output have quickly gained in importance, with a significant number of
papers dealing with the subject as discussed in references [2.46]-[2.54]. This is a consequence
of accelerated pace of developments in the multi-phase drive area in general.
The determination of appropriate PWM method for complex Matrix Converter topology,
especially for multi-phase output, is a key issue. The restrictions imposed on the PWM
strategies are that the input phases must not be short circuited and none of the output phases
should be open circuited.
The early modulation method, proposed by Alesina and Venturini as shown in reference
[2.34] for three-phase to three-phase Matrix Converter, may still be extended for multi-phase
Matrix Converter topology with little modifications, but the output voltage magnitude will
once again be limited to 50% of the input voltage magnitude. Thus it is not practical to use
this method due to the restricted output voltage magnitude. Space vector PWM approach used
for three-phase to three-phase Matrix Converter, offering higher output voltage magnitude as
given in reference [2.34], although extendable to multi-phase topology as presented in
reference [2.52], is highly complex due to large number of space vectors available. For
instance, in a three-phase to five-phase Matrix Converter total number of space vectors
generated is 215 = 32768. Nevertheless, due to constraints imposed by the switching actions,
only 35 = 243 vectors are useful. Following the analogy with five-phase three-level voltage
source inverter, the number of space vectors that can be used to generate sinusoidal output is
further limited to 93 as discussed in reference [2.52]. Nevertheless, space vector PWM for
real time processing still poses a challenge. The output voltage magnitude with space vector
PWM in a three-phase to five-phase Matrix Converter is limited to 78.86% of the input
magnitude for linear modulation range. This limit can further be raised by exploiting the
over-modulation region. No work has so far been reported on over-modulation of the multi-
phase Matrix Converter.
Carrier-based PWM scheme has been introduced for three-phase to multi-phase Matrix
Converter as discussed in references [2.50] - [2.51]. Carrier based PWM is one of the recently
developed modulation strategies for the Matrix Converter in three-phase to three-phase
topology. As in the traditional sinusoidal PWM, used in voltage source inverters, it employs
the carrier and reference signals. The implementation of this scheme is simpler compared to
the space vector PWM due to the use of conventional UP/DOWN counters, which are
embedded in most of the single chip microcontrollers. However, the carrier signal employed
in this PWM is discontinuous and thus the method is intuitively difficult to understand. In
26
addition, increasing the limit of modulation is not a trivial task. Since an offset is added in
this PWM, it is not suitable for topology where input and output neutrals needs connection.
Another simpler, compared to the space vector PWM, and modular modulation approach was
developed recently and is called ‘direct duty ratio PWM’ (DPWM). It offers the advantages
of simpler real time implementation as presented in reference [2.54]. This approach is
modular in nature and it is shown that this can be employed for any phase number or switch
number of the Matrix Converter. The major advantage of this modulation scheme is that it is
highly intuitive and flexible, and can be applied to any topology of the Matrix Converter,
since the concept is based on ‘per output phase’.
Multi-phase Matrix Converter is suitable for application in renewable energy system. In
multi-phase wind generation, a multi-phase Matrix Converter is used as an interface between
the generator and utility grid. The input side of multi-phase Matrix Converter is n-phase and
output is 3-phase since this is to be connected to utility grid as given in reference [2.55].
Simulation results are presented in reference [2.55] for the concept of higher phase input and
lower phase output. For the same Matrix Converter configuration, space vector PWM is
elaborated and five-phase is successfully transformed to three-phase and connected to the
utility grid in reference [2.56]. The major advantage of this configuration is high voltage
transfer ratio which can go up-to 104% of the input as given in reference [2.56]. Multi-phase
electric generators (that can be used in wind turbine driven configuration) possess all the
listed advantages of a multi-phase machine. The multi-phase to three-phase Matrix Converter
can give finer resolution and higher magnitude output voltages and hence better performing
overall multi-phase generation system. A dual three-phase Permanent Magnet Synchronous
Machine (two stator windings with 30° phase displacement) is considered as a possible
solution for electric ship propulsion, in conjunction with a Matrix Converter driven by a
modified direct torque control as presented in reference [2.57]. The Matrix Converter is
controlled using space vector pulse width modulation as shown in reference [2.57]. The
machine is supplied by two sets of transformers, each one providing six-phase secondary by
phase shifting the three-phase primary voltages. Each of these secondary six-phase voltages
is then applied to a separate Matrix Converter that gives a three-phase output. There are two
alternatives to the original Matrix Converter topology; called single-sided Matrix Converter
and indirect Matrix Converter, as shown in Fig. (2.2.)
27
Fig. 2.2. Single-sided Matrix Converter with unidirectional power switches as shown in reference [2.58].
Fig. 2.3. Indirect Matrix Converter with rectifier and inversion stage as shown in reference [2.58].
A single-sided multi-phase Matrix Converter that supplies unidirectional current to run a
switched reluctance motor is proposed in reference [2.59]. The topology do not use bi-
directional power switches. The converter essentially consists of a controlled rectifier applied
separately to each one of the load phases. Therefore, by adding more rectifiers, more phases
can be supplied. In essence, the topology provides easier commutation, modulation and
higher fault tolerance and modular structure.
A single sided Matrix Converter supplying brushless DC motor with fault tolerant property is
proposed in reference [2.60]. High fault tolerance and power density are claimed and results
are provided for a five-phase BLDC machine supplied by single sided Matrix Converter.
28
Application of a single-sided Matrix Converter supplying a five-phase brushless DC drive for
applications in the more electric aircraft is proposed in reference [2.61]. The real time
implementation is done using a field programmable gate array (FPGA) and hysteresis control
is employed. Hysteresis control is employed in order to maximize the system robustness and
remove the elements imposing the highest risk of failure. Experimental results are provided
for a range of operating conditions including field weakening region and single-phase open-
circuit fault.
The indirect Matrix Converter, discussed in reference [2.62], is a topology which employed
active rectifier instead of diode based rectifier. The topology uses a fully controllable rectifier
with bi-directional switches. Contrary to conventional Voltage Source Inverter, this topology
does not include a large bulky capacitor in the DC-bus, to increase the system robustness.
The clamping circuit, consisting of a diode and a smaller capacitor, use to protect from
overvoltage in the DC-link if the inductive load continues to draw current when the inverter
stage is turned off. The two stages, rectifier and inverter, are controlled independently but are
synchronized. The rectifier control aims to create maximal DC-bus voltage and minimal
distortion to the source side current, which is controlled to be in phase with the input voltage
for unity power factor operation. The inverter part of indirect Matrix Converter control relies
on accepted modulation techniques for multi-phase VSIs [2.1].
2.5 Summary
This chapter present the state-of-the art in the development of multi-phase motor drive
system. Advantages of multi-phase motor drive over three-phase drive system are
investigated and the literatures are cited where such advantages are highlighted. The
developments in the control algorithms of multi-phase power converters are addressed. The
multi-phase power converters can be AC-DC, DC-AC and AC-AC. The focus of the research
is AC-AC converter and hence literature related to AC-AC converter is mostly cited. At first
conventional three-phase input to three-phase output Matrix Converter is taken up for
discussion and the relevant literature are cited. This is followed by the discussion on three-
phase input and multi-phase output Matrix Converter is elaborated. Some recent works are
reported in the literature and the findings are summarized.
29
References:
[2.1] E. Levi, “Multi-phase electric machines for variable-speed applications,” IEEE Trans. on Industrial Electronics, vol. 55, no. 5, 2008, pp. 1893-1909.
[2.2] E. Levi, R. Bojoi, F. Profumo, H.A. Toliyat, S. Williamson, “Multi-phase induction motor drives – A technology status review,” IET – Electric Power Applications, vol. 1, no. 4, 2007, pp. 489-516.
[2.3] J.W. Kelly, E.G. Strangas, J.M. Miller, “Multi-phase inverter analysis,” Proc. IEEE Int. Electric Machines and Drives Conf. IEMDC, Cambridge, MA, 2001, pp. 147-155.
[2.4] O. Ojo, G. Dong, “Generalized discontinuous carrier-based PWM modulation scheme for multi-phase converter-machine systems,” Proc. IEEE Ind. Appl. Soc. Annual Meeting IAS, Hong Kong, 2005, CD-ROM paper 38P3.
[2.5] A. Iqbal, E. Levi, M. Jones, S.N. Vukosavic, “Generalised sinusoidal PWM with harmonic injection for multi-phase VSIs,” Proc. IEEE Power Elec. Spec. Conf. PESC, Jeju, Korea, 2006, pp. 2871-2877.
[2.6] M.J. Duran, E. Levi, “Multi-dimensional approach to multi-phase space vector pulse width modulation,” Proc. IEEE Ind. Elec. Soc. Annual Meeting IECON, Paris, France, 2006, pp. 2103-2108.
[2.7] A. Iqbal, E. Levi, “Space vector PWM techniques for sinusoidal output voltage generation with a five-phase voltage source inverter,” Electric Power Components and Systems, vol. 34, no. 2, 2006, pp. 119-140.
[2.8] J.W. Kelly, E.G. Strangas, J.M. Miller, “Multi-phase space vector pulse width modulation,” IEEE Trans. on Energy Conversion, vol. 18, no. 2, 2003, pp. 259-264.
[2.9] H.M. Ryu, J.H. Kim, S.K. Sul, “Analysis of multi-phase space vector pulse-width modulation based on multiple d-q spaces concept,” IEEE Trans. on Power Electronics, vol. 20, no. 6, 2005, pp. 1364-1371.
[2.10] S. Williamson, A.C. Smith, “Pulsating torque and losses in multi-phase induction machines,” IEEE Trans. Ind. Appl., vol. 39, no. 4, Jul./Aug. 2003, pp. 986–993.
[2.11] H. Weh, U. Schroder, “Static inverter concepts for multi-phase machines with square-wave current-field distribution,” in Proc. EPE, Brussels, Belgium, 1985, pp. 1.147–1.152.
[2.12] H.A. Toliyat, M.M. Rahimian, T.A. Lipo, “Analysis and modeling of five phase converters for adjustable speed drive applications,” in Proc. EPE, Brighton, U.K., 1993, pp. 194–199.
[2.13] H. A. Toliyat, T.A Lipo, “Analysis of concentrated winding induction machines for adjustable speed drive applications-experimental results,” IEEE Trans. Energy Conv., vol. 9, Dec. 1994, pp. 695–700.
[2.14] H.A. Toliyat, S.P. Waikar, T.A. Lipo, “Analysis and simulation of five-phase synchronous reluctance machines including third harmonic of air gap MMF,” IEEE Trans. Ind. Appl., vol. 34, no. 2, Mar./Apr. 1998, pp. 332–339.
[2.15] R.O.C. Lyra, T.A. Lipo, “Torque density improvement in a six-phase induction motor with third harmonic current injection,” IEEE Trans. Ind. Appl., vol. 38, no. 5, Sep./Oct. 2002, pp. 1351–1360.
[2.16] L. Lillo, L. Emprimgham, P.W. Wheeler, S. Khwan-on, C. Gereda, M.N. Othman, X. Huang, “Multi-phase power converter drive for fault-tolerant machine development in Aerospace applications”, IEEE Transaction on Industrial Electronics, vol. 57, no. 2, pp. 575-583, 2010.
30
[2.17] X. L. Alberti, N. bianchi, “Experimental test of Dual Three-Phase Induction Motor Under Faulty Operation Condition,” IEEE Transactions On Industrial Electronics, Vol. 59, no. 5,PP. 2041-2048, 2012.
[2.18] X. hang , A. Goodman, C. Gerada, Y. Fang, Q. Lu, “Design Of a Five Phase Brushless DC Motor For a Safety Critical Aerospace Application,” IEEE Transactions On Industrial Electronics, Vol. 59, no. 9,PP. 3532-3541, 2012.
[2.19] A. Mohammedpour, L. Parsa. “A Unified Fault tolerant current control Approach For Five Phase PM Motors with Trapezoidal Back EMF Under Different Stator Winding Connections,” IEEE Transactions On Power Electronics, Vol. 28, no. 7,PP. 3517-3527, 2013.
[2.20] Y. Zhao, T.A. Lipo. “Modelling And Control of a Multi Phase Induction Machine with Structural Unbalance Part I, Machine Modelling and Multi-dimensional Current Regulation,” IEEE Transactions On Energy Conversion, Vol. 11, no. 3,PP. 578-584, 1996.
[2.21] J-P. Martin, F. Meibody-Tabar, B. Davat. “Multi-phase Permanente Magnet Synchronous Machine Supplied By VSIs, Working Under Fault Conditions,” IEEE Industry Application Conferences, Vol. 3,PP. 1710-1717, 2000.
[2.22] R. Kaininezhad, B. Nahed-Mobarakeh, L. Baghli, F. Betin, G.A Capolino, “Modelling And Control Of Six Phase Symmetrical Induction Machine Under Fault Condition Due to Open Phase,” IEEE Transactions On Industrial Electronics, Vol. 55, no.5,PP. 1966-1977, 2008.
[2.23] J.R. Fu, T.A. Lipo, “Disturbance Free Operations Of a Multi-Phase current Regulated Motor Drive With An Opened Phase.” IEEE Transaction on Industry Applications, Vol. 30, no.5. PP. 1267-1274, 1994.
[2.24] H.M Ryu, J.W. Kim, S.K. Sul, “Synchronous-Frame Current Control of Multi Phase Synchronous Motor Under Asymmetric Fault Condition Due To Open Phases.” .” IEEE Transaction on Industry Applications, Vol. 42, no.4. PP. 1062-1274, 1994.
[2.25] A. Tani, M. Mengoni, L. Zarrii, G. Serra, D. Cassadei, “Control of Multi-phase Induction Motor with an Odd Number of Phases Under Open circuit Phase Faults.” IEEE Transactions On Power Electronics, Vol. 27, no. 2,PP. 565-572, 2012.
[2.26] S. Dwari, L. Parsa, “Fault tolerant Control of Five Phase Permanent Magnet Motors with Trapezoidal Back EMF.” IEEE Transactions on Industrial Electronics, Vol. 58, no. 2, PP. 476-485, 2012.
[2.27] F. Barrero , J. Prieto, E. Levi, R, Gregor , S. Toral , M. Duran, M. Jones “An Enhanced Predictive Current Control Method for Asymmetrical Six Phase Motor Drives.” IEEE Transactions On Industrial Electronics, Vol. 58, no. 8, PP. 3242-3252, 2011.
[2.28] J.R. Fu, T.A. Lipo, “Disturbance-free operation of a multi-phase current-regulated motor drive with an opened phase,” IEEE Trans. Ind. Appl., vol. 30, no. 5, Sep./Oct. 1994, pp. 1267–1274.
[2.29] T.M. Jahns, “Improved reliability in solid-state AC drives by means of multiple independent phase-drive units,” IEEE Trans. Ind. Appl., vol. IA-16, no. 3, May/Jun. 1980, pp. 321–331.
[2.30] M.P. Kazmierkowski, R. Krishnan, F. Blaabjerg, “Control in Power Electronics-Selected Problems”, Academic Press, UK, ISSN: 0-12-402772-5.
[2.31] P.W. Wheeler, J. Rodriguez, Jon C. Clare, L. Empringham, A. Weinstein, “Matrix Converters: A Technology Review”, IEEE Trans. On Ind. Elect. vol. 49, no. 2, April, 2002, pp. 276-288.
31
[2.32] J. Rodriguez, M. Rivera, J. W. Kolar, P. W. Wheeler, "A review of control and modulation methods for Matrix Converters," IEEE Trans. On Ind. Elec., vol. 59, no. 1, pp. 58-70, 2012.
[2.33] M. Venturini, “A new sine wave in, sine wave out, conversion technique eliminates reactive elements,” in Proceedings of Power-con 7, San Diego, CA, 1980, pp. E3-1-E3-15.
[2.34] A. Alesina, M. Venturini, “Solid state power conversion: A Fourier analysis approach to generalised transformer synthesis” IEEE Trans. Circuit System, vol. 28, no. CS-4, pp. 319-330, April 1981.
[2.35] P.D. Ziogas, S.I. Khan, M.H. Rashid, “Some Improved Forced Commutated Cyclo-converters Structure”, IEEE Transactions on Industry Applications, vol. IA-21, No 5,Sept./Oct. 1985, pp. 1242-1253.
[2.36] P.D. Ziogas, S.I. Khan, M.H. Rashid, “Analysis and Design of Forced Commutated Cyclo-converter Structures with Improved Transfer Characteristics,” IEEE Transactions on Industry Applications, vol. IE-33, No. 3, August 1986, pp. 271-280.
[2.37] A. Alesina, M. Venturnin, “Analysis and design of optimum amplitude nine-switch direct AC-AC converters”, IEEE Trans. Power Elect. vol. PE-4, no. 1, pp. 101-112, 1989.
[2.38] D. Casadei, G. Grandi, G. Serra, A. Tani, “Space vector control of Matrix Converters with unity power factor and sinusoidal input/output waveforms”, Proc. EPE Conf., vol. 7, pp. 170-175, 1993.
[2.39] L.Huber, D. Borojevic, “Space vector modulated three-phase to three-phase Matrix Converter with input power factor correction”, IEEE Trans. Ind. Appl. Vol. 31, no. 6, pp. 1234-1246, Nov./Dec. 1995.
[2.40] B. Wang, G. Venkataramanan, “A carrier-based PWM algorithm for indirect Matrix Converters,” in Proc. IEEE-PESC 2006, pp. 2780–2787, 2006.
[2.41] Young-Doo Yoon and Seung-Ki Sul, “Carrier-based modulation technique for Matrix Converter,” IEEE Trans. Power Elect., vol. 21, no. 6, pp. 1691–1703, November 2006.
[2.42] Poh C. Loh, R. Rong, F. Blaabjerg, P. Wang, “Digital carrier Modulation and Sampling Issues of Matrix Converter”, IEEE Trans. On Power Elect., vol. 24, no. 7, July 2009.
[2.43] P. Tenti, L. Malesani, L. Rossetto, “Optimum control of N-input K-output Matrix Converters,” IEEE Trans. on Power Electronics, vol. 7, no. 4, 1992, pp. 707-713.
[2.44] X. Huang, K. Bradley, A. Goodman, C. Gerada, P. Wheeler, J. Clare, C. Whitley, “Fault-tolerant brushless DC motor drive for electro-hydrostatic actuation system in aerospace application,” IEEE Industry Application Society Annual Meeting IAS, Tampa, FL, 2006, pp. 473-480.
[2.45] J. Szczepanik, “Multi-phase Matrix Converter for power systems application,” Proc. Int. Symp. on Power Electronics, Electrical Drives, Automation and Motion SPEEDAM, Ischia, Italy, 2008, pp. 772-777.
[2.46] J. Szczepanik, T. Sienko, “New control strategy for multi-phase Matrix Converter,” IEEE Int. Conf. on Systems Engineering, Las Vegas, Nevada, 2008, pp. 121-126.
[2.47] J. Szczepanik, T. Sienko, “A new concept of application of multi-phase Matrix Converter in power systems,” IEEE EUROCON – The Int. Conf. on “Computer as a Tool”, Warsaw, Poland, 2007, pp. 1535-1540.
[2.48] J. Szczepanik, T. Sienko, “Control scheme for a multi-phase Matrix Converter,” IEEE EUROCON, Saint-Petersburg, Russia, 2009, pp. 545-551.
32
[2.49] X. Huang, K. Bradley, A. Goodman, C. Gerada, P. Wheeler, J. Clare, C. Whitley, “Fault-tolerant analysis of multi-phase single sided matrix converter for brushless DC drives,” IEEE Int. Symp. on Industrial Electronics, Vigo, Spain, 2007, pp. 3168-3173.
[2.50] Sk. M. Ahmed, A. Iqbal, H. Abu-rub, M.R. Khan, “Carrier based PWM technique for a novel three-to-seven phase Matrix Converter,” Int. Conf. on Electrical Machines ICEM, Rome, Italy, 2010, CD-ROM.
[2.51] A. Iqbal, Sk. M. Ahmed, H. Abu-Rub, M.R. Khan, “Carrier based PWM technique for a novel three-to-five phase Matrix Converter,” Power Conversion and Intelligent Motion PCIM, Nuremberg, Germany, 2010, pp. 998-1003.
[2.52] Sk. M. Ahmed, A. Iqbal, H. Abu-Rub, M.R. Khan, “Space vector PWM technique for a novel three-to-five phase Matrix Converter,” IEEE Energy Conversion Conf. and Exhibition, Atlanta, GA, 2010, pp. 1875-1880.
[2.53] M. Mihret, M. Abreham, O. Ojo, S. Karugaba, “Modulation schemes for five-phase to three-phase AC-AC Matrix Converter,” IEEE Energy Conversion Conf. and Exhibition, Atlanta, GA, 2010, pp. 1887-1893.
[2.54] A. Iqbal, Sk. M. Ahmed, H. Abu-Rub, “Generalised duty ratio based pulse width modulation technique for a three-to-k phase Matrix Converter, IEEE Trans. on Industrial Electronics, under publication.
[2.55] O. Ojo, M. Abreham, S. Karugaba, O. A. Komolafe, "Carrier-based modulation of non-square multi-phase AC-AC Matrix Converters," in Proc. IEEE Int. Symp. on Ind. Elec. ISIE, Bari, Italy, pp. 2141-2146, 2010.
[2.56] O. Abdel-Rahim, H. Abu-Rub, A. Iqbal, “Five-to-Three Phase Direct Matrix Converter with Model Predictive Control”, 4th Int. conf. on Power Engineering, Energy and Electrical drives, CD-ROM paper, 13-17 May, 2013, Istanbul, Turkey.
[2.57] G. Yi, Y. Xuekui, "Research on Matrix Converter control multi-phase PMSM for all electric ship," in Proc. Int. Conf. on Electrical and Control Engineering ICECE, Yichang, China, pp. 3120-3123, 2011.
[2.58] E. Levi, N. Bodo, O. Dordevic, M. Jones, “Recent advances in power electronic converter control for multi-phase drive system”, Proc. WEMDCD, 2013.
[2.59] A. S. Goodman, K. J. Bradley, P. W. Wheeler, “Evaluation of the single sided Matrix Converter driven switched reluctance motor,” in Proc. IEEE Ind. Appl. Annual Meeting IAS, Seattle, WA, pp. 1847-1851, 2004.
[2.60] H. Xiaoyan, K. Bradley, A. Goodman, C. Gerada, P. Wheeler, J. Clare, C. Whitley, "Fault-tolerance analysis of multi-phase single sided Matrix Converter for brushless DC drives," in Proc. IEEE Int. Symp. On Ind. Elec. ISIE, Vigo, Spain, pp. 3168-3173, 2007.
[2.61] H. Xiaoyan, A. Goodman, C. Gerada, F. Youtong, L. Qinfen, "A single sided Matrix Converter drive for a brushless DC motor in aero-space applications," IEEE Trans. on Ind. Elec., vol. 59, no. 9, pp.3542-3552, 2012.
[2.62] Li Gang, “PWM Algorithams for indeirect Matrix Converter”, Proc. 7th international power electronics and motion control conference (IPEMC) , vol.3, pp. 1713-1717, 2012
33
Chapter 3 Over-View of Modelling and Control of
three-phase by three-phase Matrix-Converter
3.1 Introduction
In general, conventional Matrix Converter topologies consist of three phase input to single-or
three-phase output. The complexity of the Matrix Converter circuit configuration makes the
phase control and the determination of suitable modulation strategies a difficult task. Two
different mathematical approaches have been considered in the past to face this problem,
namely the Modulation Duty-Cycle Matrix(MDCM) or Alesina-Venturuni approach (scalar
method) and the Space Vector Modulation (SVM) approach. In recent literature two different
latest modulation techniques are introduced namely-Carrier based PWM technique and Direct
Duty-ratio based PWM technique (DDPWM). This chapter discusses the conventional Matrix
Converter topology (three-phase input and three-phase output) and different modulation
techniques for the control of Matrix Converter.
3.2 Control Algorithms for Matrix Converter
There are several different open-loop and closed-loop control methods suitable for Matrix
Converters. The open-loop methods can be considered as pure modulation strategies if load
control strategies are ignored. In addition, a closed-loop control system does not usually
control the converter itself, but rather the load, so that they also contain a modulator. Thus
modulation is one of the key issues with Matrix Converters (MC). Before the modulation
techniques are discussed in detail, the basic connection of Matrix Converter is explained.
In Fig. 3.1 subscripts d and r in the IGBTs label stand for direct and reverse respectively and
they refer to the output current flow direction, which is assumed to be direct or positive when
it is from the input to the output. With reference to Fig. 3.1, when the output phase,
accordingly to the main control algorithm, has to be commutated from one input phase to
another, two rules must be respected by any commutation strategy:
34
Fig. 3.1. Three phase (input) to single phase (output) basic connection.
i) The commutation does not have to cause a short circuit between the two input
phases, because the consequent high circulating current might destroy the
switches;
ii) The commutation does not have to cause an interruption of the output current
because the consequent overvoltage might likely destroy the switches.
To fulfil these requirements some knowledge of the commutation conditions is mandatory.
In order to carry out a safe commutation, the voltage between the involved bi-directional
switches or the output current must be measured.
These information are required in order to determine the proper sequence of the devices
switching state combinations that does not lead to the hazard either of a short circuit or of an
overvoltage and provides the safe commutation of the output current. This is the common
operating principle of all the commutation strategies that have been proposed in literature.
3.3 Modulation Duty Cycle Matrix Approach
The basic scheme of three-phase Matrix Converters has been represented in Fig. 3.2. The
switching behaviour of the converter generates discontinuous output voltage waveforms.
Assuming inductive loads connected at the output side leads to continuous output current
35
waveforms. In these operating conditions, the instantaneous power balance equation, applied
at the input and output side of an ideal converter, leads to discontinuous input currents. The
presence of capacitors at the input side is required to ensure continuous input voltage
waveforms. In order to analyse the modulation strategies, an opportune converter model is
introduced, which is valid considering ideal switches and a switching frequency much higher
than input and output frequencies. Under these assumptions, the higher frequency
components of the variables can be neglected, and the input/output quantities are represented
by their average values over a cycle period CT .
AC
AC
AC
Isb
VAVB VC
Va
Vb
VC
Ia
Ib
Ic
Input Filter
Isa
Isc
S11 S12 S13
S21 S22 S23
S31 S32 S33
3 Phase Input
3 Phase Output
L
L
L
C C C
Fig. 3.2. General topology of three phase to three phase Matrix Converter.
The input/output relationships of voltages and currents are related to the states of the nine
switches, and can be written in matrix form as shown in reference [3.1]
1 11 12 13 1
2 21 22 23 2
3 31 32 33 3
o i
o i
o i
v m m m v
v m m m v
v m m m v
(3.1)
1 11 12 13 1
2 21 22 23 2
3 31 32 33 3
i o
i o
i o
i m m m i
i m m m i
i m m m i
(3.2)
with 10 hkm , h=1,2,3 k=1,2,3. (3.3)
36
The variables hkm are the duty-cycles of the nine switches hkS and can be represented by the
duty-cycle matrix xm . In order to prevent short-circuit on the input side and ensure
uninterrupted load current flow, these duty-cycles must satisfy the three following constraint
conditions:
11 12 13
21 22 23
31 32 33
1
1
1
m m m
m m m
m m m
(3.4)
The determination of any modulation strategy for the Matrix Converter, can be formulated as
the problem of determining, in each cycle period, the duty-cycle matrix that satisfies the
input-output voltage relationships of equation (3.1),the required instantaneous input power
factor, and the constraint conditions from the equations (3.3)-(3.4). The solution of this
problem represents a hard task and is not unique, as documented by the different solutions
proposed in literature. It should be noted that in order to completely determine the
modulation strategy it is necessary to define the switching pattern, which is the commutation
sequence of the nine switches. The use of different switching patterns for the same duty-cycle
matrix leads to a different behaviour in terms of number of switch commutations and ripple
of input and output quantities.
A. Alesina-Venturini 1981 (AV method)
A first solution, obtained by using the duty-cycle matrix approach, has been proposed in
reference [3.2]. This strategy allows the control of the output voltages and input power factor,
and can be summarized in the following equation, valid for unity input power factor (αi = βi).
1 2 21 2 cos ( 1) cos ( 1)
3 3 3hk o om q h k
(3.5)
Assuming balanced supply voltages and balanced output conditions, the maximum value of
the voltage transfer ratio q is 0.5. This low value represents the major drawback of this
modulation strategy. The allocation of the switch states within a cycle period is not unique
and different switching patterns lead to different input-output ripple performance. A typical
double-sided switching pattern usually adopted is represented schematically in Fig. 3.3. It is
37
possible to see by using this modulation technique, 12 switch commutations occur in each
cycle period (a commutation takes place when the value of h or k in hkm changes).
B. Alesina-Venturini 1989 (Optimum AV method)
In order to improve the performance of the previous modulation strategy in terms of
maximum voltage transfer ratio, a second solution has been presented in reference [3.3]. In
this case the modulation law can be described by the following relationship:
2 ( 1)1 2 cos
3
cos 3 cos 32 11. cos
3 3 6 2 3
2 2 2cos 4 1 cos 2 1
3 33 3
i
o ihk o
i i
kq
hm
q k k
(3.6)
In particular, the solution given in equation (3.6) is valid for unity input power factor.
Fig. 3.3. Double-sided switching pattern in a cycle period Tp.
For unity power. factor (αi = βi), and the maximum voltage transfer ratio q is 0.866. It should
be noted that in a complete solution, valid for values of the input power factor different from
unity, has been derived. The corresponding expressions for hkm are very complex and require
the knowledge of the output power factor.
38
3.4 Space Vector Control Approach
The Space Vector Approach is based on the instantaneous space vector representation of
input and output voltages and currents. Among the 27 possible switching configurations
available in three-phase Matrix Converters, 21 only can be usefully employed in the SVM
algorithm, and can be represented as shown in Tab. 3.1.The first 18 switching configurations
determine an output voltage vector ov and an input current vector ii , having fixed directions,
as represented in Figs. 3.4(a) and (b), and is referred to as “active configurations”. The
magnitude of these vectors depends upon the instantaneous values of the input line-to-line
voltages and output line currents respectively. The last 3 switching configurations determine
zero input current and output voltage vectors and is referred to as “zero configurations”. The
remaining 6 switching configurations have each output phase connected to a different input
phase. In this case the output voltage and input current vectors have variable directions and
cannot be usefully used to synthesize the reference vectors as shown in references [3.4-3.6].
A. SVM Technique
The SVM algorithm for Matrix Converters presented in this paragraph has the inherent
capability to achieve the full control of both output voltage vector and instantaneous input
current displacement angle as presented in reference [3.1]. At any sampling instant, the
output voltage vector ov and the input current displacement angle i are known as reference
quantities (Figs. 3.4 (a) and 3.4 (b)). The input line-to-neutral voltage vector iv is imposed
by the source voltages and is known by measurements. Then, the control of i can be
achieved controlling the phase angle i of the input current vector. In principle, the SVM
algorithm is based on the selection of 4 active configurations that are applied for suitable time
intervals within each cycle periodpT . The zero configurations are applied to complete
pT .
In order to explain the modulation algorithm, reference will be made to Figs. 3.5 (a) and (b),
where ov and ii are assumed both lying in sector 1, without missing the generality of the
analysis. The reference voltage vector ov is resolved into the components 'ov and "ov along
the two adjacent vector directions. The 'ov component can be synthesised using two voltage
vectors having the same direction of 'ov . Among the six possible switching configurations
39
(±7, ±8, ±9), the ones that allow also the modulation of the input current direction must be
selected. It is verified that this constraint allows the elimination of two switching
configurations (+8 and -8 in this case). The remaining four, can be assumed to apply the
positive switching configurations (+7 and +9). The meaning of this assumption will be
discussed later in this paragraph. With similar considerations the switching configurations
required to synthesize the "ov component can be selected (+1 and +3).
Table 3.1. SWITCHING CONFIGURATIONS USED IN THE SVM ALGORITHM.
a. b.
40
Fig. 3.4. Direction of the output line-to neutral voltage vectors generated by the active configurations.
a. b. Fig. 3.4a. Output voltage vectors modulation principle. Fig. 3.4(b) - Input current vectors
modulation principle.
Using the same procedure, it is possible to determine the four switching configurations
related to any possible combination of output voltage and input current sectors, leading to the
results summarized in Tab. 3.2. Four symbols (I, II, III, IV) are also introduced in the last row
of Table. 3.2 to identify the four general switching configurations, valid for any combination
of input and output sectors. Now it is possible to write, in a general form, the four basic
equations of the SVM algorithm, which satisfy, at the same time, the requirements of the
reference output voltage vector and input current displacement angle. With reference to the
output voltage vector, the two following equations can be written:
Table 3.2. Selection of the switching configurations for each combination of output voltage
and input current sectors.
41
1 /3 /3
1 /3
2' cos (3.7)
33
2" cos (3.8)
33
v
v
j KI I II IIo o o o o
j KIII III IV IVo o o o o
v v v v e
v v v v e
With reference to the input current displacement angle, two equations are obtained by
imposing to the vectors IIIIo
IIo vv and IVIV
oIIIIII
o vv to have the direction defined by
i . This can be achieved by imposing a null-value to the two vectors component along the
direction perpendicular to ije (i.e. ijje ), leading to
1 /3. 0 ii j KjI I II IIo ov v je e
(3.9)
1 /3
. 0 ii j KjIII III IV IVo ov v je e
(3.10)
In equations (3.7)-(3.10) o
and i
are the output voltage and input current phase angle
measured with respect to the bisecting line of the corresponding sector, and differ from o
and i
according to the output voltage and input current sectors. In these equations the
following angle limits apply
66
,66
io
(3.11)
I , II , III , IV are the duty-cycles (i.e. I = pI Tt / ) of the 4 switching configurations,
Kv=1,2,..,6 represents the output voltage sector and Ki=1,2,…,6 represents the input current
42
sector. 0Iv , 0
IIv , 0IIIv , 0
IVv are the output voltage vectors associated respectively with the
switching configurations I, II, III, IV given in Tab. 3.2. The same formalism is used for the
input current vectors.
Solving equations (3.7)-(3.10) with respect to the duty-cycles, after some tedious
manipulations, leads to the following relationships as seen from reference [3.7]:
i
ooKKI qiv
cos3
cos3
cos.
3
21
(3.12)
i
ooKKII qiv
cos3
cos3
cos.
3
21 1
(3.13)
i
ooKKIII qiv
cos3
cos3
cos.
3
21 1
(3.14)
i
ooKKIV qiv
cos3
cos3
cos.
3
21
(3.15)
Equations (3.12) - (3.15) have a general validity and can be applied for any combination of
output voltage sector vK and input current sector iK .It should be noted that, for any sector
combinations, two of the duty cycles calculated by equations (3.12)-(3.15) assume negative
values. This is due to the assumption made of using only the positive switching
configurations in writing the basic equations (3.7)-(3.10). A negative value of the duty-cycle
means that the corresponding negative switching configuration has to be selected instead of
the positive one. Furthermore, for the feasibility of the control strategy, the sum of the
absolute values of the four duty-cycles must be lower than unity.
1 IVIIIIII (3.16)
The zero configurations are applied to complete the cycle period. By introducing equations
(3.12)-(3.15) in equation (3.16), after some manipulations, leads to the following equation.
43
io
iq
coscos
cos
2
3 (3.17)
Equation (3.17) represents, at any instant, the theoretical maximum voltage transfer ratio,
which is dependent on the output voltage and input current phase angles and the displacement
angle of the input current vector. It is useful to note that, in the particular case of balanced
supply voltages and balanced output voltages, the maximum voltage transfer ratio occurs
when equation (3.17) is a minimum (i.e. when i
cos and o
cos are equal to 1), leading to
iq cos2
3 (3.18)
Assuming unity input power factor, equation (3.18) gives the well-known maximum voltage
transfer ratio of Matrix Converters 0.866.Using the SVM technique, the switching pattern is
defined by the switching configuration sequence. With reference to the particular case of
output voltage vector lying in sector 1 and input current vector lying in sector 1, the
switching configurations selected are, in general, 01, 02, 03, +1, -3, -7, +9. It can be verified
that there is only one switching configuration sequence characterized by only one switch
commutation for each switching configuration change, that is 03, -3, +9, 01, -7, +1, 02. The
corresponding general double-sided switching pattern is shown in Fig. 3.5.
The use of the three zero configuration leads to 12 switch commutations in each cycle period.
It should be noted that the possibility to select the duty-cycles of three zero configurations
gives two degree of freedom, being 9371030201 1 . These two degree of
freedom can be utilized to define different switching patterns, characterized by different
behaviour in terms of ripple of the input and output quantities. In particular, the two degree of
freedom might be utilized to eliminate one or two zero configurations, affecting also the
number of commutations in each cycle period.
44
Fig. 3.5. Double-sided switching pattern in a cycle period Tp [3.7].
3.7 Carrier-Based PWM Approach
A. Carrier Based Control Strategy
Considering a balanced three-phase system, the input voltages can be given as
)3/4cos(
)3/2cos(
)cos(
tVv
tVv
tVv
b
b
a
(3.19)
The duty ratios should be so chosen that the output voltage remains independent of input
frequency. In other words the three-phase balanced input voltages can be considered to be in
stationary reference frame and the output voltage can be considered to be in synchronous
reference frame, so the that input frequency term will be absent in the output voltage as
shown in references [3.8-3.10]. Hence the duty ratios cAbAaA danddd , are chosen as
)3/4cos(
),3/2cos(
),cos(
tkd
tkd
tkd
cA
bA
aA
(3.20)
Therefore the phase A output voltage can be obtained by using the above duty ratios as
45
)]3/4cos()3/4cos(
)3/2cos()3/2cos()cos()[cos(
tt
ttttVkv AA
(3.21)
In general equation (3.21) can be written as:
)cos(2
3 Vkv AA
(3.22)
In equation (3.22), )cos( term indicates that the output voltage is affected by the choice of
and later it will be explained that the input power factor depends on . Thus, the output
voltage AV is independent of the input frequency and only depends on the amplitude V
of the
input voltage and Ak is a time-varying signal with the desired output frequency o . The 3-
phase reference output voltages can be represented as:
(3.23)
Where kB and kC are the reference output voltage modulating signals for the output phases B
and C respectively. Therefore, from equation (3.22), the output voltage in phase-A is:
)cos()cos(2
3tVkv oAA
(3.24)
B. Application of Offset Duty Ratios
In the discussions, mentioned in the section (3.7. A), duty-ratios become negative which are
not practically realizable. For the switches connected to output phase-A, at any instant, the
condition 1 ,,0 cAbAaA ddd should be valid. Therefore, offset duty ratios should to be added
to the existing duty-ratios, so that the net resultant duty-ratios of individual switches are
always positive. Furthermore, the offset duty-ratios should be added equally to all the output
phases to ensure that the resultant output voltage vector produced by the offset duty ratios is
null in the load. That is, the offset duty-ratios can only add the common-mode voltages in the
output. Considering the case of output phase-A as shown in reference [3.9]:
(3.25)
0)3/4cos()3/2cos()cos(, tktktkddd AAAcAbAaA
)3/4cos(
),3/2cos(
),cos(
tmk
tmk
tmk
oC
oB
oA
46
To cancel the negative components from individual duty ratios absolute value of the duty-
ratios are added. Thus the minimum individual offset duty ratios should be
)3/4cos(5.0)()3/2cos(5.0)(,)cos(5.0)( ttDandttDttD cba
respectively. The effective duty ratios are )( ),(),( tDdtDdtDd ccAbbAaaA . The same
holds good for the input phases b and c. The net duty ratio )(tDd aaA should be
accommodated within a range of 0 to 1.
Thus 1)(0 tDd aaA , i.e 1)cos()cos(0 tktk AA . This implies that in the
worst case 1.20 Ak The maximum value of Ak or in other words ‘m’ in equation (3.23) is
equal to 0.5. Hence the offset duty-ratios (Fig. 3.5) corresponding to the three input phases
are chosen as
)3/4cos(5.0)(
,)3/2cos(5.0)(
,)cos(5.0)(
ttD
ttD
ttD
oc
ob
oa
(3.26)
Fig. 3.6. Modified offset duty ratios for all input phases.
Thus the modified duty ratios for output phase-A are
)3/4cos()(
),3/2cos()(
),cos()(
tmtDd
tmtDd
tmtDd
occA
obbA
oaaA
(3.27)
0 0.01 0.02 0.03 0.04 0.050
0.2
0.4
0.6
Time
Mag
nitu
de
OFFSET
47
In any switching cycle the output phase has to be connected to any of the input phases. The
summation of the duty ratios in equation (3.27) must equal to unity. But the summation
)()()( tDtDtD cba is less than or equal to unity as shown in Fig. 3.5. Hence another
offset duty-ratio 3/)()()(1 tDtDtD cba is added )()(),( tandDtDtD cba in equation
(3.27). The addition of this offset duty-ratio in all switches will maintain the output voltages
and input currents unaffected. Similarly, the duty-ratios are calculated for the output phases B
and C.
If Ak , Bk and Ck are chosen to be 3-phase sinusoidal references as given in Equation (3.23),
the input voltage capability is not fully utilized for output voltage generation. To overcome
this, an additional common mode term equal to 2/),,min(),,max( CBACBA kkkkkk is
added as in the carrier-based space-vector PWM principle as implemented in two-level
inverters. Thus the amplitude of Ak , Bk and Ck can be enhanced from 0.5 to 0.57.
Thus the duty-ratios for output phase-A are modified as:
)3/4cos(]2/),,min(
),,max([3/))()()(1()(
)3/2cos(]2/),,min(
),,max([3/))()()(1()(
)cos(]2/),,min(
),,max([3/))()()(1()(
tkkk
kkkktDtDtDtDd
tkkk
kkkktDtDtDtDd
tkkk
kkkktDtDtDtDd
CBA
CBAAcbaacA
CBA
CBAAcbaabA
CBA
CBAAcbaaaA
(3.28)
3.8 Direct Duty Ratio Based PWM approach
In this section, the pulse width modulation technique is discussed based on duty ratio
calculation in conjunction with the generalised three-to three-phase topology of the proposed
Matrix Converter. The duty ratio based PWM (DPWM) is developed by using the concept of
per-phase output average over one switching period as discussed in references [3.10-3.11].
The developed scheme is modular in nature and is thus applicable to the generalised
converter circuit topology.
It is assumed that a switching period of the carrier wave consist of two sub-periods, T1
(rising slope of the triangular carrier) and T2 (falling slope of the triangular carrier). When the
carrier changes from zero to the peak value, the sub-period is called T1, and when the carrier
48
changes from peak to the zero value it is termed as sub-interval T2. The input three-phase
sinusoidal waveform can assume different values at different instants of times. The maximum
among the three input signals is termed as Max, the medium amplitude among three input
signals is termed as Mid and the smallest magnitude is represented as Min. During interval T1
(positive slope of the carrier), the line-to-line voltage between Max and Min (
,,,, CBACBA vvvMinvvvMax ) phases is used for the calculation of duty ratio and no
consideration is given to the medium amplitude of input signal. The output voltage should
initially follow the Max signal of the input and then should follow the Min signal of the input.
During interval T2, two line voltages between Max and Mid ( ,,,, CBACBA vvvMidvvvMax )
and Mid and Min ( ,,,, CBACBA vvvMinvvvMid ) is calculated first, and the largest among the
two is used for the calculation of the duty ratio. This is done to balance the volt-second
principle. Two different cases can arise in time period T2 depending upon the relative
magnitude of the input voltages. If (Max-Mid) > (Mid-Min), the output should follow Max for
certain time period and then follow Mid for certain time period. This situation is termed as
Case I. This is further explained in the next section. Similarly, if Max-Mid <Mid-Min, the
output should follow at first Mid of the input signal and then Min of the input signal and this
is termed as Case II. Thus DPWM approach uses two out of three of the line-to-line input
voltages to synthesis output voltages, and all the three input phases are utilized to conduct
current during each switching period. Case I and Case II and the generation of gating signals
are further elaborated in the next section.
i) Case-I: For condition (Max-Mid) (Mid-Min), the generation of gating pattern for kth
output phase is illustrated in Fig. 3.6 for one switching period. To generate the
pattern, at first the duty ratio cbakDk &,,1 , is calculated and then compared with
high frequency triangular carrier signal to generate the kth output phase pattern. The
gating pattern for the kth leg of the Matrix Converter is directly derived from the
output pattern. The switching pattern is drawn assuming that the Max is the phase ‘A’
of the input, Mid is the phase ‘B’ and Min is the phase ‘C’. The switching pattern
changes in accordance with the variation in the relative magnitude of the input phases.
The output follows Min of the input signal if the magnitude of the duty ratio is more
than the magnitude of the carrier and the slope of the carrier is positive. The output
follows the Max of the input signal if the magnitude of the carrier is more than the
magnitude of the duty ratio irrespective of the slope of the carrier. Finally, the output
tracks Mid if the magnitude of the carrier signal is smaller compared to the magnitude
49
of the duty ratio and the slope of the carrier is negative. Thus, the resulting output
phase voltage changes like Min→Max→Max→Mid. These transition periods are
termed as, 321 ,, kkk ttt and 4kt and these four sub-intervals can be expressed presented in
references [3.10-3.12]:
skk TDt 11
skk TDt 12 1
skk TDt 11 13 (3.29)
skk TDt 114
4321 kkkks ttttT
Where 1kD is the kth phase duty ratio value, when Case I is under consideration and is
defined by sT
T1 , which refers to the fraction of the slope of the carrier. Now, by using the
volt-second principle of PWM control, the following equation can be written:
*1 2 3 4
0
, , . , , . , , .sT
ok s ok A B C k A B C k k A B C kv T v dt Min v v v t Max v v v t t Mid v v v t (3.30)
Substituting the time intervals expressions from equation (3.29) into equation (3.30), yield
*1
0
. , , . , , 1 , ,
, , , ,
sTA B C A B C
ok ok k A B CA B C A B Cs
Min v v v Mid v v vv v dt D Max v v v
Mid v v v Max v v vT
(3.31)
Where Ts is the sampling period, okok vv ,* are the reference and actual average output voltage
of phase ‘k’, respectively and CBA vvv ,, are the input side three-phase voltages. Max, Mid and
Min refer to the maximum, medium and minimum values, Dk represents the duty ratio of the
power switch.
50
okv
*okv
1kD
sT
1kt 2kt 3kt 4kt
1T 2T
Fig. 3.7. Output and Switching pattern for kth phase in the Case I.
The duty ratio is obtained from equation (3.31) as:
,,,,
,, *
1CBACBA
okCBAk vvvMinvvvMid
vvvvMaxD
(3.32)
Where ,,,, CBACBA vvvMidvvvMax
Similarly, the duty ratios of other output phases can be obtained which can subsequently be
used for implementation of the PWM scheme.
ii) Case-II: Now considering another situation of Max-Mid <Mid-Min. The output and
the switching patterns can be derived once again following the same principle laid
down in the previous sub section. Fig. 3.7 shows the output and switching pattern for
the kth output phase. Here once again a high frequency triangular carrier wave is
compared with the duty ratio value, 2kD to generate the switching pattern. The only
difference in this case when compared to the previous one is the interval when the
magnitude of the carrier signal is greater than the magnitude of the duty ratio and the
slope is negative. Then, the output should follow Mid instead of Max. Contrary to
Case I, for this situation the output must follow Max of the input. The time intervals
321 ,, kkk ttt and 4kt are the same as in equation (3.29) and now the output phase voltage
is changed with the sequence of Min→Max→Mid→Min. The volt-second principle is
51
now applied to derive the equation for the duty ratio. The volt-second principle
equation can be written as:
*1 4 2 3
0
, , . , , . , , .sT
ok s ok A B C k k A B C k A B C kv T v dt Min v v v t t Max v v v t Mid v v v t (3.33)
Now once again substituting the time expression from equation (3.29) into equation
(3.33), one obtains:
,,,,. ,,.
,,.,,
,,.,,12
0
*
CBACBACBA
CBACBA
CBACBAk
T
oks
ok
vvvMidvvvMidvvvMax
vvvMidvvvMid
vvvMaxvvvMinDdtv
Tv
s
(3.34)
The duty ratio can now be obtained as:
,,,,.
,,. *
2CBACBA
okCBAK vvvMinvvvMid
vvvvMidD
(3.35)
The switching signals for the bi-directional power switching devices can be generated by
considering the switching states of Fig. (3.6) and of Fig. (3.7.) Depending upon the output
pattern, the gating signals are derived. If the output pattern of phase “k” is Max (or Mid, Min),
the output phase “k” is connected to the input phase whose voltage is Max (or Mid, Min).
The input voltages are at first examined for their relative magnitudes and the phases with
maximum, medium, and minimum values are determined. The information about their
relative magnitudes are given to the next computation block along with the commanded
output phase voltages. The computation block either uses equation (3.32) or equation (3.35)
to generate the duty ratios depending upon the relative magnitude of the input voltages. The
duty ratio obtained goes to the PWM block. The PWM block calculates the time sub-interval
using equation (3.29). The gating pattern is then derived accordingly and given to the Matrix
Converter.
52
okv
*okv 2kD
sT
1kt 2kt 3kt 4kt
1T 2T Fig. 3.8. Output and Switching pattern for kth phase in the Case II.
3.9 Summary
This chapter encompasses the modelling and control issues of a three-phase input and three-
phase output Matrix Converter. Modelling based on space vector approach is elaborated in
several literature which is summarized in this chapter. Control issues are also discussed in
this chapter. It is seen that the first PWM reported in the literature, called ‘scalar control’
produces only 50% output, in other words, the maximum obtainable output voltage is only
50% that of input voltage. The control approach is improved by injecting harmonic
components and the output is raised to 86% of the input value. The quality of output voltage
is further enhanced by employing space vector PWM technique. The implementation of space
vector PWM is quite complex. This is followed by the discussion on Carrier-based PWM and
direct duty ratio based PWM.
References:
[3.1] M.P. Kazmierkowski, R. Krishnan, F. Blaabjerg, “Control in Power Electronics-Selected Problems”, Academic Press, UK, 2002.
[3.2] A. Alesina, M. Venturini, “Solid state power conversion: A Fourier analysis approach to generalised transformer synthesis,” IEEE Trans. Circuit System, vol. 28, no. CS-4, pp. 319-330, April 1981.
[3.3] A. Alesina, M. Venturini, “Analysis and design of optimum-amplitude nine-switch direct AC-AC converters,” IEEE Transactions on Power Electronics., vol. 4, no. 1, pp. 101–12, Jan 1989.
53
[3.4] L. Huber, D. Borojevic, “Space vector modulator for forced commutated cycloconverters,” in Conference Record of the IEEE Industry Applications Society Annual Meeting, 1989, vol. 1, pp. 871–6.
[3.5] L. Huber, D. Borojevic, N. Burany, “Voltage space vector based pwmcontrol of forced commutated cycloconverters,” in IECON ’89. 15th Annual Conference of IEEE Industrial Electronics Society, 1989, vol. 1, pp. 106–11.
[3.6] L. Huber, D. Borojevic, “Space vector modulation with unity input power factor for forced commutated cyclo-converters,” in Conference Record of the 1991 IEEE Industry Applications Society Annual Meeting,1991, vol. 1, pp. 1032–41.
[3.7] R.A. Petrocelli, “New Modulation methods for Matrix Converters”, PhD Thesis, Manchester University, May 2002.
[3.8] Y. D. Yoon, S. K. Sul, “Carrier-Based Modulation Technique for Matrix Converter.” IEEE Trans. Power Electron, vol. 21, no. 6, pp 1691-1703, Nov. 2006.
[3.9] P. Satish, K.K. Mohapatra, N. Mohan, “Carrier-based control of Matrix Converter in linear and over-modulation region,” SCSC 2007, pp. 98-105.
[3.10] Yulong Li, Nam-Sup Choi, "Carrier Based Pulse Width Modulation for Matrix Converters," proceeding of IEEE APEC Conf., Washington DC, February, 2009.
[3.11] L. Yulong, C. Nam-Sup , H. Byung-Moon, N. Eui-Cheol, “Fault-tolerant modulation strategy with direct duty-ratio PWM for Matrix Converter-fed motor drives”, IEEE 8th International Conference on Power Electronics and ECCE Asia (ICPE & ECCE), 2011, pp. 2891 – 2898.
[3.12] Yulong Li, Nam-Sup Choi, Byung-Moon Han, Kyoung Min Kim, Buhm Lee, Jun-Hyub Park, “Direct Duty Ratio Pulse Width Modulation Method for Matrix Converter,” International Journal of Control, Automation, and Systems, vol. 6, no. 5, Oct. 2008, pp. 660-669.
54
Chapter 4 Modelling of Multi-phase Multi-motor
Drive System
4.1 Introduction
Concept for multi-motor drive systems, based on utilization of multi-phase machines and
multi-phase power converters, have been proposed almost a decade ago as specified in
reference [4.1]. Since field oriented control of any multi-phase machine requires only two
stator current components, the additional stator current components available in multi-phase
machine can be used to control other machines. It has been shown that, by connecting multi-
phase stator windings in series/parallel with an appropriate phase transposition, it is possible
to control independently all the machines with supply coming from a single multi-phase
power converter. One specific drive system, covered by this general concept, is the five-phase
series-connected two-motor drive, consisting of two five-phase machines and supplied from a
single five-phase voltage source inverter. Such topology has been analysed in a considerable
depth as discussed in references [4.1]-[4.2] and experimental verification of the existence of
control decoupling in this two-motor drive has been provided in references [4.3]-[4.6]. The
studies are based on inverter current control in the stationary reference frame, using phase
current control in conjunction with hysteresis or ramp-comparison current controllers. The
experimental rig utilizes ramp-comparison current control as shown in reference [4.3].
The control techniques developed so far for the five-phase voltage source inverter feeding
five-phase series-connected two-motor drive, once again are based on pulse width
modulation. Carrier-based sinusoidal PWM are used in reference [4.4]. A modification in the
scheme is suggested in reference [4.4] where fifth harmonic is injected in the reference
voltages. With the harmonic injection the output voltage magnitude is increased, similar to
single-motor drive. A number of space vector PWM techniques have been reported for a
multi-phase VSI (five, six, seven and nine phases) for single motor drive in references [4.7-
4.25] where attempts have been made to generate sinusoidal waveform. Considering five-
phase system there exist two orthogonal planes namely d-q and x-y. Unwanted low-order
harmonics are generated in the output of a five-phase VSI when the space vectors of x-y plane
are not eliminated completely and they result in distortion in stator current and losses in the
55
machine having sinusoidal mmf distribution. In case of concentrated winding machine, low
order harmonic currents are injected along with the fundamental to enhance the torque
production. In such cases it is desirable to produce low-order harmonic along with the
fundamental as illustrated in reference [4.27].
The use of other types of power converters such as direct AC-AC converter, back-to-back
converter is rarely used in the literature to control the multi-phase multi-motor drive system.
Use of direct Matrix Converter is elaborated further in Chapters 5-7.
A specific case of, six-phase two-motor drive is presented in reference [4.6] where a
symmetrical six-phase machine (60˚ phase displacement with single neutral point) connected
in series with a three-phase machine is described. Similar to series connection of multi-phase
machines, parallel connection are also possible with independent control of each machine and
supplied from one inverter as shown in reference [4.26]. The series-connected multi-motor
drive increases the copper losses of both machines and thus lowers the efficiency. However,
it is suggested that for special applications where the load requirement is such that all the
motors are not fully loaded simultaneously, this technique is beneficial. One such application
is identified as winder drive.
Another drive system, covered by the general concept of series and parallel connection, is the
seven-phase series-connected three-motor drive, consisting of three seven-phase machines
and supplied from a single seven-phase voltage source inverter. This drive system is reported
in references [4.27-4.28]. However, no mathematical model of the overall drive structure is
presented in the literature. The modelling of seven-phase series-connected drive system is
covered in this chapter.
A number of space vector PWM techniques have been reported for a seven-phase VSI for
single motor drive references [4.28-4.30] where attempts have been made to generate
sinusoidal waveform.
Considering a seven-phase system there exist three orthogonal planes namely d-q, x1-y1 and
x2-y2. Unwanted low-order harmonics are generated in the output of a seven-phase VSI when
the space vectors of x1-y1 and x2-y2 plane are not eliminated completely and they result in
distortion in stator current and losses in the machine having sinusoidal mmf distribution. In
case of concentrated winding machine, low order harmonic currents are injected along with
the fundamental to enhance the torque production. In such cases it is desirable to produce
low-order harmonic along with the fundamental.
This chapter elaborate the modelling of three different drive structures;
Five-phase series-connected two-motor drive
56
Six-phase series-connected two-motor drive, and
Seven-phase series-connected three-motor drive.
4.2 Modelling of five-phase series-connected two-motordrives
Block diagram of the two-motor drive systems is illustrated in Fig. 4.1.
Fig. 4.1. Five-phase series-connected two-motor drive structure.
In Fig. 4.1, the source is a five-phase power converter that can be either a direct AC-AC
converter (Matrix Converter) or a voltage source inverter, which directly feed a five-phase
machine. The five-phase machine has open-end windings, the second end of the first machine
windings are connected with appropriate phase transposition to the second five-phase
machine. The second end of the windings is shorted to form the star point. The connection
scheme is given in Table 4.1.
Table 4.1. Connectivity matrix for five-phase two motor drive as shown in reference [4.3].
Machine number
a b C d e
1 a1 b1 c1 d1 e1
2 a2 c2 e2 b2 d2
A. Modelling of series-connected five-phase two-motor drive
Due to the series connection of two stator windings according to Fig. 4.1 the following voltage and current relations can be written as shown in reference [4.9].
b
e
d
c
a
b1
e1
d1
c1
a1
b2
e2
d2
c2
a2
Source Machine 1 Machine 2
57
1 2
1 2
1 2
1 2
1 2
A as as
B bs cs
C cs es
D ds bs
E es ds
v v v
v v v
v v v
v v v
v v v
(4.1)
1 2
1 2
1 2
1 2
1 2
A as as
B bs cs
C cs es
D ds bs
E es ds
i i i
i i i
i i i
i i i
i i i
(4.2)
In a general case the two machines, although both five-phase, may be different types i.e.
induction machine or permanent magnet synchronous machine or synchronous reluctance
machine etc, and therefore, may be with different parameters. Let the index ‘1’ denote
machine 1 that is directly connected to the five-phase source (note that the source can be any
power electronic converter such as inverter or Matrix Converter) and let the index ‘2’ stand
for the second machine 2, connected after the first machine through phase transposition. It is
important to note that the modelling of the two-motor drive system is independent of the type
of power converter being used as source. MC denotes the Matrix Converter.
Voltage balance equation for the complete system can be written in a compact matrix form as
dt
iLdiRv (4.3)
where the system is of the 15th order and:
0
0
MCv
v
2
1
r
r
MC
i
i
i
i (4.4)
TEDCBA
MC
TEDCBA
MC
iiiiii
vvvvvv
(4.5)
Terdrcrbrarr
Terdrcrbrarr
iiiiii
iiiiii
222222
111111
(4.6)
58
The resistance and inductance matrices of equation (4.3) can be written as:
1 2
1
2
0 0
0 0
0 0
s s
r
r
R R
R R
R
(4.7)
1 2 1 2
1 1
2 2
' '
0
' 0
s s sr sr
rs r
rs r
L L L L
L L L
L L
(4.8)
Super script’ in equation (4.8) denotes sub-matrices of machine 2 that have been modified
through the phase transposition operation, compared to their original form. The sub-matrices
of equations (4.7)-(4.8) are all five by five matrices and are given with the following
expressions 2 / 5 :
222222
111111
222222
111111
)(
)(
rrrrrr
rrrrrr
ssssss
ssssss
RRRRRdiagR
RRRRRdiagR
RRRRRdiagR
RRRRRdiagR
(4.9)
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
cos cos 2 cos 2 cos
cos cos cos 2 cos 2
cos 2 cos cos cos 2
cos 2 cos 2 cos cos
cos cos 2 cos 2 cos
ls
ls
s ls
ls
ls
L M M M M M
M L M M M M
L M M L M M M
M M M L M M
M M M M L M
(4.10)
2 2 2 2 2 2
2 2 2 2 2 2
2 2 2 2 2 22
2 2 2 2 2 2
2 2 2 2 2 2
cos 2 cos cos cos 2
cos 2 cos 2 cos cos
' cos cos 2 cos 2 cos
cos cos cos 2 cos 2
cos 2 cos cos cos 2
ls
ls
lss
ls
ls
L M M M M M
M L M M M M
L M M L M M M
M M M L M M
M M M M L M
(4.11)
111111
111111
111111
111111
111111
1
cos2cos2coscos
coscos2cos2cos
2coscoscos2cos
2cos2coscoscos
cos2cos2coscos
MLMMMM
MMLMMM
MMMLMM
MMMMLM
MMMMML
L
lr
lr
lr
lr
lr
r
(4.12)
222222
222222
222222
222222
222222
2
cos2cos2coscos
coscos2cos2cos
2coscoscos2cos
2cos2coscoscos
cos2cos2coscos
MLMMMM
MMLMMM
MMMLMM
MMMMLM
MMMMML
L
lr
lr
lr
lr
lr
r
(4.13)
59
Tsrrs
sr
LL
ML
11
11111
11111
11111
11111
11111
11
coscos2cos2coscos
coscoscos2cos2cos
2coscoscoscos2cos
2cos2coscoscoscos
cos2cos2coscoscos
(4.14)
''
)cos(coscos2cos2cos
2cos)2cos(coscoscos
coscos)2cos(2coscos
2coscoscos)cos(2cos
cos2cos2coscoscos
'
22
22222
22222
22222
22222
22222
22
Tsrrs
sr
LL
ML
(4.15)
Expansion of equation (4.3) produces the following:
0
0
MCv
v =1 2
1
2
0 0
0 0
0 0
s s
r
r
R R
R
R
2
1
r
r
MC
i
i
i
+
22
11
2121
0'
0
''
rrs
rrs
srsrss
LL
LL
LLLL
2
1
r
r
MC
i
i
i
dt
d+
00'
00
'0
2
1
21
rs
rs
srsr
Ldt
d
Ldt
d
Ldt
dL
dt
d
2
1
r
r
MC
i
i
i
(4.16)
where:
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1 1 11
1 1 1 1 1
1 1 1 1 1
sin sin sin 2 sin 2 sin
sin sin sin sin 2 sin 2
sin 2 sin sin sin sin 2
sin 2 sin 2 sin sin sin
sin sin 2 sin 2 sin sin
sr
dL M
dt
1 1T
rs sr
d dL L
dt dt
(4.17)
and
60
2 2 2 2 2
2 2 2 2 2
2 2 2 2 2 2 22
2 2 2 2 2
2 2 2 2 2
sin sin sin 2 sin 2 sin
sin 2 sin( ) sin sin sin 2
' sin sin 2 sin( 2 ) sin sin
sin sin sin sin( 2 ) sin 2
sin 2 sin 2 sin sin sin( )
sr
dL M
dt
2 2' 'Trs sr
d dL L
dt dt
(4.18)
Torque equations of the two machines in terms of source currents and their respective rotor
currents and rotor positions are obtained as:
)sin()2sin(
)2sin(
)sin(sin
1111111
11111111111
111111111111
111
erAdrEcrDbrCarB
erBdrAcrEbrDarCerCdrBcrAbrEarD
erDdrCcrBbrAarEerEdrDcrCbrBarA
e
iiiiiiiiii
iiiiiiiiiiiiiiiiiiii
iiiiiiiiiiiiiiiiiiii
MPT
)sin()2sin(
)2sin(
)sin(sin
2222222
22222222222
222222222222
222
erAdrCcrEbrBarD
erDdrAcrCbrEarBerBdrDcrAbrCarE
erEdrBcrDbrAarCerCdrEcrBbrDarA
e
iiiiiiiiii
iiiiiiiiiiiiiiiiiiii
iiiiiiiiiiiiiiiiiiii
MPT
(4.19)
The model given with equation (4.16) and (4.19) together with equations (4.7)-(4.15) and
equations (4.17)-(4.18) constitutes the 15th order model of the complete two-motor drive with
phase transposition in the series connection of stator windings in phase-variable form. The
equations of motion are:
1 11 1
1e L
d PT T
dt J
2 22 2
2e L
d PT T
dt J
(4.20)
4.2b Model in the Rotating Reference Frame
In order to simplify the phase-domain model of Section 4.2a, the decoupling transformation
is applied. The Clark’s decoupling transformation matrixin power invarient form given in
reference [4.2] is:
1 cos cos 2 cos 3 cos 4
0 sin sin 2 sin 3 sin 42 1 cos 2 cos 4 cos 6 cos85
0 sin 2 sin 4 sin 6 sin 8
0 1/ 2 1/ 2 1/ 2 1/ 2 1/ 2
C x
y
(4.21)
61
The new variables are defined as:
ivffCf MCMC , , (4.22)
Application of equation (4.22) in conjunction with equation (4.3) can be written as follows
(the notation 1 12.5mL M and 2 22.5mL M applies further on):
Matrix Converter/stator voltage equations:
111111
11111112121
cossin
sincos
rrm
rmrmMC
mlslsMC
ssMC
iiL
piLpiLpiLLLiRRv
111111
11111112121
cossin
sincos
rrm
rmrmMC
mlslsMC
ssMC
iiL
piLpiLpiLLLiRRv
222222
22222212121
cossin
sincos
rrm
rmrmMCxmlsls
MCxss
MCx
iiL
piLpiLpiLLLiRRv
(4.23a)
222222
22222212121
sincos
cossin
rrm
rmrmMCymlsls
MCyss
MCy
iiL
piLpiLpiLLLiRRv
MColsls
MCoss
MCo piLLiRRv 2121 (4.23b)
where, dt
dp
Rotor voltage equations of machine 1:
MCMC
m
rmlrMC
mMC
mrrr
iiL
piLLpiLpiLiRv
1111
1111111111
cossin
sincos0
MCMC
m
rmlrMC
mMC
mrrr
iiL
piLLpiLpiLiRv
1111
1111111111
sincos
cossin0
(4.24a)
11111 0 xrlrxrrxr piLiRv
62
11111 0 yrlryrryr piLiRv
11111 0 orlrorror piLiRv
Rotor voltage equations of machine 2:
MC
yMCxm
rmlrMCym
MCxmrrr
iiL
piLLpiLpiLiRv
2222
2222222222
cossin
sincos0
MC
yMCxm
rmlrMCym
MCxmrrr
iiL
piLLpiLpiLiRv
2222
2222222222
sincos
cossin0
22222 0 xrlrxrrxr piLiRv (4.24b)
22212 0 yrlryrryr piLiRv
22222 0 orlrorror piLiRv
Electromagnetic torque equations of two machines are obtained as:
MC
yrMCxr
MCxr
MCyrme
MCr
MCr
MCr
MCrme
iiiiiiiiLPT
iiiiiiiiLPT
222222222
111111111
sincos
sincos
(4.25)
The torque equations of the two machines show that the torque of machine 1 entirely depends
upon the α-axis and β-axis components of source current (Matrix Converter), while the torque
developed by machine 2 is due to x-axis and y-axis components of source current (Matrix
Converter). This implies that torque of machine 1 can be controlled by controlling α-axis and
β-axis components of source current while torque of machine 2 can be controlled by
controlling x-axis and y-axis components of source current. Thus an independent control of
two machine is possible. It can be seen from equations (4.23)-(4.24) that the x-axis, y-axis
current and o-axis components of the two rotors are completely decoupled from the rest of
the system. Hence they can be omitted from further consideration. The zero-sequence
component current of stator will be zero for isolated neutral system. Hence the equation for
inverter zero-sequence component can be omitted as well. Rotor x, y, and zero-sequence
component equations, as well as the stator zero-sequence component equations are
nevertheless retained for the time being, for the sake of completeness.
63
4.2c Model in the stationary reference frame
Rotational transformation is applied to the rotor equations of both machines 1 and 2 of the
two-motor system model, to obtain the model in the stationary common reference frame. The
rotational transformation matrix of equation (4.26) is applied to the rotor of machine 1 and
equation (4.27) is applied to the rotor of machine 2. Since all the x-y components and zero-
sequence components are decoupled, rotational transformation needs to be applied to the α-
axis and β-axis components only.
1 1
1 1
1
cos sin 0 0 0
sin cos 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
D
(4.26)
2 2
2 2
2
cos sin 0 0 0
sin cos 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
D
(4.27)
The angles of transformations are 1 1 and 2 2 1 1 2 2,dt dt (the instantaneous
angular positions of the d-axis of the common reference frame with respect to the phases ‘a’
magnetic axes of the rotors) and 1 2, are the electrical angular speeds of rotation of the
machine 1 and machine 2, respectively. By omitting the x-y and zero-sequence equation for rotor windings and the zero-sequence
equation of the power source (Matrix Converter), the complete d-q model in stationary
reference frame for the two five-phase series-connected machines can be written in developed
form as:
Stator Equations:
dt
diL
dt
diLLiR
dt
diLiRv
dt
diL
dt
diLLiR
dt
diLiRv
piLiRpiLpiLLiRv
piLiRpiLpiLLiRv
qrm
INVy
mlsINVys
INVy
lsINVys
INVy
drm
INVx
mlsINVxs
INVx
lsINVxs
INVx
MCqls
MCqsqrm
INVqmls
MCqs
MCq
MCdls
MCdsdrm
MCdmls
MCds
MCd
2222211
2222211
2211111
2211111
)(
)(
)(
)(
(4.28)
Rotor Equations:
Of machine-1
64
111111111111
111111111111
0
0
drmlrMCdmqrmlr
MCqmqrrqr
qrmlrMCqmdrmlr
MCdmdrrdr
iLLiLpiLLpiLiRv
iLLiLpiLLpiLiRv
(4.29)
Of machine-2
222222222212
222222222122
0
0
drmlrMCxmqrmlr
MCymqrrqr
qrmlrMCymdrmlr
MCxmdrrdr
iLLiLpiLLpiLiRv
iLLiLpiLLpiLiRv
(4.30)
In addition, the individual stator voltage equations can be derived as;
MCqls
MCqsys
MCdls
MCdsxs
qrmMCymls
MCysqs
drmMCxmls
MCxsds
MCyls
MCysys
MCxls
MCxsxs
qrmMCqmls
MCqsqs
drmMCdmls
MCdsds
iLiRv
piLiRv
iLpiLLiRv
iLpiLLiRv
piLiRv
piLiRv
piLpiLLiRv
piLpiLLiRv
222
222
222222
222222
111
111
111111
111111
)(
)(
)(
)(
(4.31)
Torque equations of the two machines become:
22222
11111
qrMCx
MCydrme
qrMCd
MCqdrme
iiiiLPT
iiiiLPT
(4.32)
When the phase variable equations are transformed using decoupling matrix, three sets of
equations are obtained, namely d-q, x-y and zero sequence. In single five-phase motor drives,
the d-q components are involved in actual electromagnetic energy conversion while the x-y
components increase the thermal loading of the machine. However, the extra set of current
components (x-y) available in a five-phase system is effectively utilised in independently
controlling an additional five-phase machine when the stator windings of two five-phase
machines are connected in series and are supplied from a single five-phase power source.
Reference currents generated by two independent vector controllers, are summed up as per
the transposition rules and are supplied to the series-connected five-phase machines.
65
4.2d Simulation Results
Simulation is done for one five-phase IM fed using ideal voltage source , one machine is run
at 40 Hz and second machine run at 30 Hz with constant v/f. The results are shown in Fig.
4.2. The independence of control is seen from the simulation results. The results also validate
the model of derived in the previous section.
a.
b.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-2
0
2
4
6
8
10
12
Time (s)
Tor
que
IM1
(Nm
)
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-1
0
1
2
3
4
5
6
7
8
9
Time (s)
Tor
que
IM2
(Nm
)
66
c.
d.
Fig. 4.2. Response of Five-phase Two-motor drive supplied by ideal voltage source, a. Torque of machine 1, b. Torque of machine 2, Speeds of machine 1 and 2, d. Rotor fluxes for
machine 1 and 2.
4.3 Modeling of a Six-phase Series-connected Two-motor Drive System
Connection diagram for series connection of stator windings of a six-phase and a three-phase
machine is shown in Fig. 4.3 as presented in reference [4.3].
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
200
400
600
800
1000
1200
1400
Time (s)
Spe
ed I
M1
and
IM2
(rpm
)
IM1
IM2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.2
0.4
0.6
0.8
1
1.2
1.4
Time (s)
Rot
or f
lux
IM1
and
IM2
(Wb)
IM1
IM2
67
Ai
Bi
Fi
Ei
Di
Ci
Fig. 4.3. Connection diagram for series connection of a six-phase and a three-phase machine.
4.3a Phase Variable Model
This section develops the model of the complete six-phase two-motor drive system by
considering the three-phase machine as a virtual six-phase machine. Let the parameters and
variables of the six-phase machine be identified with index 1, while index 2 applies to the
three-phase machine. Since the system of Fig. 4.3 is six-phase, it is convenient to represent
the three-phase machine as a ‘virtual’ six-phase machine, meaning that the spatial
displacement stays at 60° and the phases a2, b2, c2 of the three-phase machine are actually
phases a2, c2, e2 of the virtual six-phase machine with spatial displacements of 120°. Hence
the three-phase machine can be represented as a virtual six-phase machine with the following
set of equations:
222 2
2 22 22
sss s
s srs rs
dv R i
dt
L i L i
(4.33)
222 2
2 22 22
rrr r
r rsr sr
dv R i
dt
L i L i
(4.34)
where
68
2 2 22
2 2 22
2 2 22
0 0 0
0 0 0
0 0 0
T
as cs ess
T
as cs ess
T
as cs ess
v v v v
i i i i
(4.35)
2 2 22
2 2 22
2 2 22
0 0 0
0 0 0
0 0 0
T
ar cr err
T
ar cr err
T
ar cr err
v v v v
i i i i
(4.36)
The matrices of stator and rotor inductances are given with ( 62 ):
2 2 2 2
2 2 2 22
2 2 2 2
0 cos 2 0 cos 4 0
0 0 0 0 0 0
cos 4 0 0 cos 2 0
0 0 0 0 0 0
cos 2 0 cos 4 0 0
0 0 0 0 0 0
ls
lss
ls
L M M M
M L M ML
M M L M
(4.37)
2 2 2 2
2 2 2 22
2 2 2 2
0 cos 2 0 cos 4 0
0 0 0 0 0 0
cos 4 0 0 cos 2 0
0 0 0 0 0 0
cos 2 0 cos 4 0 0
0 0 0 0 0 0
lr
lrr
lr
L M M M
M L M ML
M M L M
(4.38)
while mutual inductance matrices between stator and rotor windings are:
2 2 2
2 2 222
2 2 2
2 2
cos 0 cos 4 0 cos 2 0
0 0 0 0 0 0
cos 2 0 cos 0 cos 4 0
0 0 0 0 0 0
cos 4 0 cos 2 0 cos 0
0 0 0 0 0 0
sr
Trs sr
L M
L L
(4.39)
Resistance matrices are:
2 2 22
2 2 22
( 0 0 0)
( 0 0 0)s s ss
r r rr
R diag R R R
R diag R R R
(4.40)
and the machine torque is:
2 2 2 2 2 2 2 2 2 2 2 2 2 2
2 2 2
2 2 2 2 2 2 2
sin sin( 4 )
sin( 2 )
as ar cs cr es er es ar as cr cs er
e
cs ar es cr as er
i i i i i i i i i i i iT P M
i i i i i i
(4.41)
69
Correlation between machine voltages and source voltages is given with Fig. 4.3, where the
phases of the three-phase machine are now labelled as a2, c2, e2. Hence
1 2 1 2 1 2
1 2 1 2 1 2
A a a B b c C c e
D d a E e c F f e
v v v v v v v v v
v v v v v v v v v
(4.42)
Correlation between machine currents and source currents is the following:
MC
F
E
D
C
B
A
fs
es
ds
cs
bs
as
s i
i
i
i
i
i
i
i
i
i
i
i
i
i
1
1
1
1
1
1
1 (4.43)
0
0
0
2
2
2
2
2
2
2
FC
EB
DA
fs
es
ds
cs
bs
as
s
ii
ii
ii
i
i
i
i
i
i
i (4.44)
Voltage equations of the complete two-motor system are formulated in terms of source
currents and voltages. The system is of the 18th order, since the three-phase machine is
represented as a virtual six-phase machine. However, three rotor equations will be redundant
(equations for rotor phases b, d, f). The complete set of voltage equations can be written as:
2
1
2'
11
'1
'21
2
1
2
1
'21
0
0
00
00
00
0
0
r
r
INV
rrs
rrs
srsss
r
r
MC
r
r
ssMC
i
i
i
LL
LL
LLLL
p
i
i
i
R
R
RRv
(4.45)
All the inductance, resistance and null sub-matrices are of six by six order. Primed sub-
matrices are those that have been modified, with respect to their original form as equation
(4.45), in the process of series connection of two machines through the phase transposition
operation. These sub-matrices are equal to:
70
2 2
2 2
2 22
2 2
2 2
2 2
0 0 0 0
0 0 0 0
0 0 0 0'
0 0 0 0
0 0 0 0
0 0 0 0
s s
s s
s ss
s s
s s
s s
R R
R R
R RR
R R
R R
R R
(4.46)
2 2 2 2 2 2 2 2
2 2 2 2 2 2 2 2
2 2 2 2 2 2 2 22
2 2 2 2 2 2 2 2
2 2 2 2 2
cos 2 cos 4 cos 2 cos 4
cos 4 cos 2 cos 4 cos 2
cos 2 cos 4 cos 2 cos 4'
cos 2 cos 4 cos 2 cos 4
cos 4 cos 2 cos 4
ls ls
ls ls
ls lss
ls ls
ls
L M M M L M M M
M L M M M L M M
M M L M M M L ML
L M M M L M M M
M L M M M
2 2 2
2 2 2 2 2 2 2 2
cos 2
cos 2 cos 4 cos 2 cos 4ls
ls ls
L M M
M M L M M M L M
(4.47)
2 2 2
2 2 2
2 2 222
2 2 2
2 2 2
2 2 2
2 2
cos 0 cos 4 0 cos 2 0
cos 2 0 cos 0 cos 4 0
cos 4 0 cos 2 0 cos 0'
cos 0 cos 4 0 cos 2 0
cos 2 0 cos 0 cos 4 0
cos 4 0 cos 2 0 cos 0
' '
sr
T
rs sr
L M
L L
(4.48)
Torque equations of the two machines can be given in terms of inverter currents as:
2 2 2 2
2 2 2 2 2 2 2
2 2 2 2
( ) ( ) ( ) sin
( ) ( ) ( ) sin( 4 )
( ) ( ) ( ) sin( 2 )
A D ar B E cr C F er
e C F ar A D cr B E er
B E ar C F cr A D er
i i i i i i i i i
T P M i i i i i i i i i
i i i i i i i i i
(4.49a)
1 1 1 1 1 1 1
1 1 1 1 1 1 1
1 1 1 1 1 1 1
1 1 1
1 1 1 1 1
sin
sin( 5 )
sin( 4 )
A ar B br C cr D dr E er F fr
F ar A br B cr C dr D er E fr
E ar F br A cr B dr C er D fr
e
D ar E br F cr A dr B er C fr
i i i i i i i i i i i i
i i i i i i i i i i i i
i i i i i i i i i i i iT PM
i i i i i i i i i i i i
1 1
1 1 1 1 1 1 1
1 1 1 1 1 1 1
sin( 3 )
sin( 2 )
sin( )
C ar D br E cr F dr A er B fr
B ar C br D cr E dr F er A fr
i i i i i i i i i i i i
i i i i i i i i i i i i
(4.49b)
4.3b Model in the Rotating reference frame
To obtain orthogonal form of the model, the following Clark’s decoupling transformation
matrices in power invariant form are applied to the phase variable model:
71
6
1 cos cos 2 cos3 cos 4 cos5
0 sin sin 2 sin 3 sin 4 sin 5
1 cos 2 cos 4 cos 6 cos8 cos1020 sin 2 sin 4 sin 6 sin 8 sin106
0 1 2 1 2 1 2 1 2 1 2 1 2
0 1 2 1 2 1 2 1 2 1 2 1 2
xC
y
(4.50a)
(3)
1 cos 2 cos 42
0 sin 2 sin 43
1 2 1 2 1 2
C
(4.50b)
Axis components of source output phase voltages are:
F
E
D
C
B
A
MC
MC
MCy
MCx
MC
MC
MC
v
v
v
v
v
v
C
v
v
v
v
v
v
v 6
0
0
(4.51)
Application of equation (4.50a) - (4.50b) in conjunction with equation (4.42) produces:
10
2010
21
21
1
1
21
21
21
21
21
21
6
0
02
2
2
s
ss
sys
sxs
s
s
ef
ce
ad
ec
cb
aa
MC
MC
MCy
MCx
MC
MC
MC
v
vv
vv
vv
v
v
vv
vv
vv
vv
vv
vv
C
v
v
v
v
v
v
v
(4.52)
The complete decoupled model of the six-phase two-motor drive system. Matrix Converter-
stator voltage equations are obtained as:
1 1 1 1 1 1 1cos sinMC MC MCs s m r rv R i L pi pL i i
1 1 1 1 1 1 1sin cosMC MC MCs s m r rv R i L pi pL i i
1 1 2 2 2 2 2 2 22 2 2 cos sinMC MC MC MC MCx s x ls x s x s x m r rv R i L pi R i L p i L p i i
1 1 2 2 2 2 2 2 22 2 2 sin cosMC MC INV MC MCy s y ls y s y s y m r rv R i L pi R i L p i L p i i
0 1 0 1 0 2 0 2 02 2 2MC MC MC MC MCs ls s lsv R i L pi R i L p i
0 1 0 1 0MC MC MC
s lsv R i L pi (4.53
72
where 1 13mL M , 2 21.5mL M . Relationship between inverter current axis components and
axis components of stator currents of the two machines is found in the following form:
1
1
1 2
1 2
0 0 1 0 2
0 0 1
2
2
2
MCs
MCs
MCx xs s
MCy ys s
MCs s
MCs
i i
i i
i i i
i i i
i i i
i i
(4.54)
Rotor voltage equations of the six-phase machine result in the form:
1 1 1 1 1 1 1
1 1 1 1 1 1 1
1 1
0 cos sin
0 sin cos
0 p 1, 1, 0 1, 0 1
IMC MCr r r r m
MC MCr r r r m
r k lr k
R i L pi pL i i
R i L pi pL i i
R i L i k xr yr r r
(4.55)
While rotor voltage equations of the three-phase machine become:
2 2 2 2 2 2 2
2 2 2 2 2 2 2
2 0 2 2 0 2
0 2 cos sin
0 2 sin cos
0
MC MCr r r r m x y
MC MCr r r r m x y
r r lr r
R i L pi p L i i
R i L pi p L i i
R i L pi
(4.56)
Torque equations of the two machines are:
1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2
cos sin
2 cos sin
MC MC MC MCe m r r r r
MC MC MC MCe m r y r x r x r y
T P L i i i i i i i i
T P L i i i i i i i i
(4.57)
4.3c Transformation of the model into the stationary common reference frame
Rotational transformation is applied to rotor equations of the two machines. The obtained
model in the stationary common reference frame is described as following. Since the x-y and
zero-sequence components of the rotor voltages and currents are zero, the complete model in
the stationary reference frame for the six-phase and the three-phase series connected
machines consists of ten differential equations. The developed form of inverter voltage
equations is:
73
1 1 1 1 1
1 1 1 1 1
1 1 2 2 2 2 2
1 1 2 2 2 2 2
0 1
( )
( )
2 2 ( ) 2
2 2 ( ) 2
MC MC MCs ls m m dr
MC MC MCs ls m m qr
MC MC MC MC MCx s x ls x s x ls m x m dr
MC IMC MC MC MCy s y ls y s y ls m y m qr
MCs
v R i L L pi L pi
v R i L L pi L pi
v R i L pi R i L L p i L pi
v R i L pi R i L L p i L pi
v R
0 1 0 2 0 2 0
0 1 0 1 0
2 2 2MC MC MC MCls s ls
MC MC MCs ls
i L pi R i L p i
v R i L pi
(4.58)
Rotor voltage equations of the six-phase machine are:
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
0
0
MC MCr dr m lr m dr m lr m qr
MC MCr qr m lr m qr m lr m dr
R i L pi L L pi L i L L i
R i L pi L L pi L i L L i
(4.59)
and rotor voltage equations of the three-phase machine are:
2 2 2 2 2 2 2 2 2 2 2
2 2 2 2 2 2 2 2 2 2 2
0 2 2
0 2 2
MC MCr dr m x lr m dr m y lr m qr
MC MCr qr m y lr m qr m x lr m dr
R i L pi L L pi L i L L i
R i L pi L L pi L i L L i
(4.60)
Torque equations of the two machines are
1 1 1 1 1
2 2 2 2 22
MC MCe m dr q d qr
MC MCe m dr y x qr
T PL i i i i
T P L i i i i
(4.61)
The mathemetical equations developed in this section is verified assuming ideal sinusoidal
source. The independence of control is observed from the Fig. 4.4.
a.
0 0.2 0.4 0.6 0.8 1
-5
0
5
10
15
20
25
30
35
Time (s)
Tor
que
IM1(
Nm
)
74
b.
c.
Fig. 4.4. Response of Six-phase Two-motor drive supplied by ideal voltage source.
4.4 Modelling of A Seven-Phase Series-connected Three-Motor Drive System
A seven-phase system, when transformed, has four orthogonal components in a seven
dimensional space. However, for field oriented control only two components are needed
(component with 90 phase shift).Hence in a seven-phase system, at least three orthogonal
pairs are available to control realizable three field oriented controllers. Thus, if the stator
windings of three seven-phase machines are connected in series (without any change in the
rotor part), all these three machines can be controlled independently as presented in reference
[4.6]. To obtain decoupled control of all the three machines, it is important that the d-q
current of one machine becomes the x-y of the other two machines. Similarly, the x-y of the
first machine becomes the d-q of the others. With such arrangement, the component that will
0 0.2 0.4 0.6 0.8 1
-2
0
2
4
6
8
10
Time(s)
Tor
que IM
2 (N
m)
0 0.2 0.4 0.6 0.8 1
0
200
400
600
800
1000
1200
1400
1600
Time (s)
Spee
d IM
1 an
d IM
2 (r
pm)
IM2
IM1
75
produce the rotating magnetic field in one machine will not produce any rotating magnetic
field in the other two machines. This component is then responsible for the torque production
while the other component will be limited by the leakage impedance, will not produce any
torque. This is achieved by proper phase transposition, between the stator winding
connections.
The power supply to three series-connected machine is given by a seven-phase voltage or
current controlled PWM inverters or Matrix Converter. In other words, it is possible to
independently implement field oriented control of three seven-phase series-connected
machines using a single power converter source. In this chapter seven-phase induction
machines with spatial displacement of 2π/7 between the phases are considered. Although the
concept of independence of control of series-connected machine do not pose any constraints
on the type of machine being used. The connection diagram showing three seven-phase
machines with stator winding connected in series and supplied by a single power source is
given in Fig. 4.5. The stator winding of the three machines are connected in series while the
three rotors can independently take three different loads. The three series-connected machines
can run under identical loading conditions or they are independent of taking up any loads.
A
B
C
D
E
F
G
a1 a2 a3
b1 b2 b3
c1 c2 c3
d1 d2 d3
e1 e2 e3
f 1 f 2 f 3
g1 g2 g3
MACHINE 1 (S1) MACHINE 2 (S2) MACHINE 3 (S3)
VaS1 VaS2 VaS3
VbS1 VbS2 VbS3
VcS1 VcS2 VcS3
VdS1 VdS2 VdS3
VeS1 VeS2 VeS3
VfS1 VfS2 VfS3
VgS1 VgS2 VgS3
Fig. 4.5. Seven Phase Series-Connected Three-Motor system.
The phase transposition rule of three machine connections is given in Table 4.2. The
machines are labelled as M1, M2 and M3 (in column), the phases of the power source are
shown in the rows.
76
Table 4.2. Connectivity Matrix.
4.4a Phase Variable Model
Phase variable model of two seven-phase induction machines connected in series according
to Fig. 4.5 is developed in state space form. Power source (Matrix Converter) voltage of each
phase after series connection as shown in Fig 4.5 can be as they are determined in an
appropriate summation of stator phase voltages of individual machine with respect to the
stator phase connection.
321
321
321
321
321
321
321
eSfSgSG
bSdSfSF
fSbSeSE
cSgSdSD
gSeScSC
dScSbSB
aSaSaSA
VVVV
VVVV
VVVV
VVVV
VVVV
VVVV
VVVV
(4.62)
The current through each phase winding is:
321
321
321
321
321
321
321
eSfSgSG
bSdSfSF
fSbSeSE
cSgSdSD
gSeScSC
dScSbSB
aSaSaSA
IIII
IIII
IIII
IIII
IIII
IIII
IIII
(4.63)
After adding angular displacement of frequency, the supply voltage will become:
Machines A B C D E F G M1 a1 b1 C1 d1 e1 f1 g1
M2 a2 c2 e2 g2 b2 d2 f2
M3 a3 g3 f3 e3 d3 c3 b3
77
)78)(2()7
10)(2()712)(2(
)72)(2()7
6)(2()710)(2(
)710)(2()7
2)(2()78)(2(
)74)(2()7
12)(2()76)(2(
)712)(2()7
8)(2()74)(2(
)76)(2()7
4)(2()72)(2(
))(2())(2())(2(
332211
332211
332211
332211
332211
332211
332211
tfSinVtfSinVtfSinVV
tfSinVtfSinVtfSinVV
tfSinVtfSinVtfSinVV
tfSinVtfSinVtfSinVV
tfSinVtfSinVtfSinVV
tfSinVtfSinVtfSinVV
tfSinVtfSinVtfSinVV
G
F
E
D
C
B
A
(4.64)
Each phase to neutral voltage of power source is indicated with capital letters. The connected
three seven phase machines all parameters may be different. The denomination 1 indicates
the first machines parameters 2 and 3 for other two. First machine is directly connected to
seven phase inverter and phase transposed voltage is supplied to next machines.
The voltage equation in matrix form for the complete system can be represented as:
dt
diRV
or
dt
idLiRV
(4.65)
where :
3
2
1
r
r
r
MC
v
v
v
v
V (4.66)
3
2
1
321
000
000
000
000
r
r
r
sss
R
R
R
RRR
R (4.67a)
3
2
1
r
r
r
MC
i
i
i
i
i (4.67b)
78
31
3
21
2
11
13
121
13
121
00
00
00
rrs
rrs
rrs
srsrsrsss
LL
LL
LL
LLLLLL
L (4.68)
The superscript 1 indicates the modified inductance matrices according to the phase
transportation rule. The sub inductance matrix for the inductance of stator and rotor can be
represent in seven by seven matrix format as:
ggsgfsgesgdsgcsgbsgas
fgsffsfesfdsfcsfbsfas
egsefseesedsecsebseas
dgsdfsdesddsdcsdbsdas
cgscfscescdsccscbscas
bgsbfsbesbdsbcsbbsbas
agsafsaesadsacsabsaas
s
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
L (4.69)
ggrgfrgergdrgcrgbrgar
fgrffrferfdrfcrfbrfar
egrefreeredrecrebrear
dgrdfrderddrdcrdbrdar
cgrcfrcercdrccrcbrcar
bgrbfrberbdrbcrbbrbar
agrafraeradracrabraar
r
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
LLLLLLL
L (4.70)
The phase difference between each phase is denoted as 7
2 and ‘M’ is mutual
inductance between windings. After adding this values for all three stator and rotor windings
MLMCosMCosMCosMCosMCosMCos
MCosMLMCosMCosMCosMCosMCos
MCosMCosMLMCosMCosMCosMCos
MCosMCosMCosMLMCosMCosMCos
MCosMCosMCosMCosMLMCosMCos
MCosMCosMCosMCosMCosMLMCos
MCosMCosMCosMCosMCosMCosML
L
ls
ls
ls
ls
ls
ls
ls
s
2332
2332
2233
3223
3322
2332
2332
(4.70a)
79
11111111
11111111
11111111
11111111
11111111
11111111
1111111
1
2332
2332
2233
3223
3322
2332
23321
MLCosMCosMCosMCosMCosMCosM
CosMMLCosMCosMCosMCosMCosM
CosMCosMMLCosMCosMCosMCosM
CosMCosMCosMMLCosMCosMCosM
CosMCosMCosMCosMMLCosMCosM
CosMCosMCosMCosMCosMMLCosM
CosMCosMCosMCosMCosMCosMML
L
ls
ls
ls
ls
ls
ls
ls
s
(4.70b)
22222222
22222222
22222222
22222222
22222222
22222222
22222222
2
2332
2233
3223
3223
3223
3322
2332
MLCosMCosMCosMCosMCosMCosM
CosMMLCosMCosMCosMCosMCosM
CosMCosMMLCosMCosMCosMCosM
CosMCosMCosMMLCosMCosMCosM
CosMCosMCosMCosMMLCosMCosM
CosMCosMCosMCosMCosMMLCosM
CosMCosMCosMCosMCosMCosMML
L
ls
ls
ls
ls
ls
ls
ls
s
(4.70c)
33333333
33333333
33333333
33333333
33333333
33333333
33333333
3
2332
2233
3223
3223
3223
3322
2332
MLCosMCosMCosMCosMCosMCosM
CosMMLCosMCosMCosMCosMCosM
CosMCosMMLCosMCosMCosMCosM
CosMCosMCosMMLCosMCosMCosM
CosMCosMCosMCosMMLCosMCosM
CosMCosMCosMCosMCosMMLCosM
CosMCosMCosMCosMCosMCosMML
L
ls
ls
ls
ls
ls
ls
ls
s
(4.70d)
11111111
11111111
11111111
11111111
11111111
11111111
1111111
1
2332
2332
2233
3223
3322
2332
23321
MLCosMCosMCosMCosMCosMCosM
CosMMLCosMCosMCosMCosMCosM
CosMCosMMLCosMCosMCosMCosM
CosMCosMCosMMLCosMCosMCosM
CosMCosMCosMCosMMLCosMCosM
CosMCosMCosMCosMCosMMLCosM
CosMCosMCosMCosMCosMCosMML
L
lr
lr
lr
lr
lr
lr
lr
r
(4.70e)
80
22222222
22222222
22222222
22222222
22222222
22222222
22222222
2
2332
2332
2233
3223
3322
2332
2332
MLCosMCosMCosMCosMCosMCosM
CosMMLCosMCosMCosMCosMCosM
CosMCosMMLCosMCosMCosMCosM
CosMCosMCosMMLCosMCosMCosM
CosMCosMCosMCosMMLCosMCosM
CosMCosMCosMCosMCosMMLCosM
CosMCosMCosMCosMCosMCosMML
L
lr
lr
lr
lr
lr
lr
lr
r
(4.70f)
33333333
33333333
33333333
33333333
33333333
33333333
33333333
3
2332
2332
2233
3223
3322
2332
2332
MLCosMCosMCosMCosMCosMCosM
CosMMLCosMCosMCosMCosMCosM
CosMCosMMLCosMCosMCosMCosM
CosMCosMCosMMLCosMCosMCosM
CosMCosMCosMCosMMLCosMCosM
CosMCosMCosMCosMCosMMLCosM
CosMCosMCosMCosMCosMCosMML
L
lr
lr
lr
lr
lr
lr
lr
r
(4.70g)
1111111
1111111
1111111
1111111
1111111
1111111
1111111
11
)()2()3()3()2()(
)()()2()3()3()2(
)2()()()2()3()3(
)3()2()()()2()3(
)3()3()2()()()2(
)2()3()3()2()()(
)()2()3()3()2()(
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
ML sr
(4.70h)
)()()2()3()3()2(
)3()2()()()2()3(
)()3()3()2()()(
)()()2()3()2()(
)2()()()2()3()3(
)3()3()2()()()2(
)()2()3()3()2()(
2222222
2222222
2222222
2222222
2222222
2222222
1112222
21
2
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
ML sr
(4.70i)
)3()()()2()3()3(
)()3()3()2()()(
)()()2()3()3()2(
)3()3()2()()()2(
)3()()2()3()3()(
)3()2()()()2()3(
)()2()3()3()2()(
3333332
3333323
3323333
3333333
3333333
2333333
3333333
31
3
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
CosCosCosCosCosCosCos
ML sr
(4.70j)
81
While putting all the matrices in the equation (4.65), the following is obtained:
3
2
1
1
1
1
111
3
2
1
31
21
11
111
13
11
3
2
1
3
2
1
321
000
000
000
0
00
00
00
000
000
000
000
0
0
0
3
2
32
3
2
322
r
r
r
MC
rs
sr
r
r
r
MC
r
r
rrs
srss
r
r
r
MC
r
r
r
sssMC
i
i
i
i
Ldt
d
Ldt
d
Ldt
d
Ldt
dL
dt
dL
dt
d
i
i
i
i
dt
d
LL
LL
LL
LLLLLL
i
i
i
i
R
R
R
RRRV
rs
rs
srsr
rs
rs
srsrs
(4.71)
where:
1111111
1111111
1111111
1111111
1111111
1111111
1111111
111
)()2()3()3()2()(
)()()2()3()3()2(
)2()()()2()3()3(
)3()2()()()2()3(
)3()3()2()()()2(
)2()3()3()2()()(
)()2()3()3()2()(
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
MLdtd
sr
(4.71a)
)()()2()3()3()2(
)3()2()()()2()3(
)()3()3()2()()(
)()()2()3()2()(
)2()()()2()3()3(
)3()3()2()()()2(
)()2()3()3()2()(
2222222
2222222
2222222
2222222
2222222
2222222
2222222
221
2
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
MLdtd
sr
(4.71b)
)3()()()2()3()3(
)()3()3()2()()(
)()()2()3()3()2(
)3()3()2()()()2(
)3()()2()3()3()(
)3()2()()()2()3(
)()2()3()3()2()(
3333332
3333323
3323333
3333333
3333333
2333333
3333333
331
3
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
SinSinSinSinSinSinSin
MLdtd
sr
(4.71b)
111
1 srT
rs Ldt
dL
dt
d
211
2 srT
rs Ldt
dL
dt
d (4.72)
311
3 srT
rs Ldt
dL
dt
d
82
Torque equations of the seven-phase induction machine in terms of source currents and their
respective rotor currents and rotor positions are:
)()
)2()
)3()
)3()
)2()
sin)
)
11111111
11111111
11111111
11111111
11111111
11111111
11111111
111
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
iiiiiiiiiiiiii
Siniiiiiiiiiiiiii
MPT
grAfrGerFdrEcrDbrCarB
grBfrAerGdrFcrEbrDarC
grCfrBerAdrGcrFbrEarD
grDfrCerBdrAcrGbrFarE
grEfrDerCdrBcrAbrGarF
grFfrEerDdrCcrBbrAarG
grGfrFerEdrDcrCbrBarA
e
(4.73)
)()
)2()
)3()
)3()
)2()
)()
)
22222222
22222222
22222222
22222222
22222222
22222222
22222222
222
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
MPT
grAfrGerFdrEcrDbrCarB
grBfrAerGdrFcrEbrDarC
grCfrBerAdrGcrFbrEarD
grDfrCerBdrAcrGbrFarE
grEfrDerCdrBcrAbrGarF
grFfrEerDdrCcrBbrAarG
grGfrFerEdrDcrCbrBarA
e
(4.74)
)()
)2()
)3()
)3()
)2()
)()
)
33333333
33333333
33333333
33333333
33333333
33333333
33333333
333
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
Siniiiiiiiiiiiiii
MPT
grAfrGerFdrEcrDbrCarB
grBfrAerGdrFcrEbrDarC
grCfrBerAdrGcrFbrEarD
grDfrCerBdrAcrGbrFarE
grEfrDerCdrBcrAbrGarF
grFfrEerDdrCcrBbrAarG
grGfrFerEdrDcrCbrBarA
e
(4.75)
Equation of motion:
83
)(
)(
)(
333
33
222
22
111
11
Le
Le
Le
TTJ
p
dt
d
TTJ
p
dt
d
TTJ
p
dt
d
(4.76)
4.4b Model in the rotating reference frame
The decoupling transformation matrix to transfer from phase variable form to power invariant
form of seven phase system can be written as:
21
21
21
21
21
21
21
1815129630
181512`9631
121086420
121086421
654320
654321
0
7
2
2
2
1
1
SinSinSinSinSinSin
CosCosCosCosCosCos
SinSinSinSinSinSin
CosCosCosCosCosCos
SinSinSinSinSinSin
CosCosCosCosCosCos
y
x
y
x
C (4.77)
The first two rows of the decoupling matrix is defined all variables that will lead to
fundamental flux and torque production (α, β components). The last row defined the zero
sequence components and all x-y components are noted in middle.
The new variables are defined with transformation matrix:
332211
33
22
11
33
22
11
, , ,
,
rrrrrrInvMC
rr
rr
rr
MCMC
rr
rr
rrMCMC
CCCC
iCi
iCi
iCi
iCi
vCV
vCV
vCVvCV
(4.78)
The superscripts, MC stand for Matrix Converter quantities; r1, r2 and r3 refer to the rotor of
machine-1, machine-2 and machine-3.
84
Application of the new variables in to voltage equation will yield following voltage
equations:
31
21
11
1
31
21
11
11
0
0
0
r
r
r
inv
r
r
r
MCMC
iC
iC
iC
iC
dtdL
iC
iC
iC
iC
R
VC
V
(4.79)
That is:
31
21
11
1
1
1
1
111
31
21
11
1
31
21
11
111
13
11
31
21
11
1
3
2
1
3211
000
000
000
0
00
00
00
000
000
000
000
0
0
0
3
2
32
3
2
322
r
r
r
MC
rs
sr
r
r
r
MC
r
r
rrs
srss
r
r
r
MC
r
r
r
sssMC
iC
iC
iC
iC
Ldt
d
Ldt
d
Ldt
d
Ldt
dL
dt
dL
dt
d
iC
iC
iC
iC
dtd
LL
LL
LL
LLLLLL
iC
iC
iC
iC
R
R
R
RRRVC
rs
rs
srsr
rs
rs
srsrs
(4.80)
Multiply both sides with the decoupling transformation matrixC and separate terms of three
machines will give:
1 11 2 3 1 11 1
1 11 1 1
2 22
3 3 3
0 0 0 0 0
0 0 00 0 00 0 00 0 0 0 00 0 00 0 0 0 0
MC MCMC
s s s s srr rr rs r
r rr
r r r
i iR R R C L C CL CV
i iR d C L C CL CR dti i
Ri i
1 1 1 12 2 3 31 1
1 1 112 2 3 3
2 2
3 3
0 0 0 0
0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
MC MC
s sr s srr r
rs r rs rr r
r r
i iC L C C L C C L C C L C
i id dC L C CL C C L C CL Cdt dti i
i i
(4.81)
This equation further written as:
85
1 1
1 2 31 1
1 11
2 22
3 3 3
0 00 0 0
0 0 00 0 00 0 00 0 0 0 00 0 00 0 0 0 0
MC MCs sr
MCs s s
r rrs r
rr r
r
r r r
i iL LR R RVi iR d L L
R dti iR
i i
1 1 1 1
1 1
1 1
2 2 3 3
1 1
2 22 2
3 33 3
0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 00 0
0 00 0 0 0
MC MCs sr s sr
r r
r rrs r
rs rr r
i iL L L L
i id d
dt dti iL L
L Li i
(4.82a)
After taking individual matrices and solving, the top row belongs to the stator and the
next three rows are that of three rotors.
1 1
1 2 31 1
1 11
2 22
3 3 3
0 00 0 0
0 0 00 0 00 0 00 0 0 0 00 0 00 0 0 0 0
MC MCs sr
MCs s s
r rrs r
rr r
r
r r r
i iL LR R RVi iR d L L
R dti iR
i i
1 1 1 1
1 1
1 1
2 2 3 3
1 1
2 22 2
3 33 3
0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 00 0
0 00 0 0 0
MC MCs sr s sr
r r
r rrs r
rs rr r
i iL L L L
i id d
dt dti iL L
L Li i
(4.82b)
Individual sub matrices can be written as:
1
1
1
1
1
11
11
11
000000
000000
000000
000000
000000
000005.20
0000005.2
ls
ls
ls
ls
ls
ls
ls
s
L
L
L
L
L
ML
ML
CLC (4.83)
86
1
1
1
1
1
11
11
11
000000
000000
000000
000000
000000
000005.20
0000005.2
lr
lr
lr
lr
lr
lr
lr
r
L
L
L
L
L
ML
ML
CLC (4.84)
0000000
0000000
0000000
0000000
0000000
00000
00000
5.2
11
11
11
1
CosSin
SinCos
MCLC sr (4.85)
Tsrrs CLCCLC 11
11
(4.86)
2
2
2
22
22
2
2
12
000000
000000
000000
0005.2000
00005.200
000000
000000
ls
ls
ls
ls
ls
ls
ls
s
L
L
L
ML
ML
L
L
CLC (4.87)
2
2
2
22
22
2
2
12
000000
000000
000000
0005.2000
00005.200
000000
000000
lr
lr
lr
lr
lr
lr
lr
r
L
L
L
ML
ML
L
L
CLC (4.88)
87
0000000
0000000
0000000
00000
00000
0000000
0000000
5.2 22
22
21
2
CosSin
SinCos
MCLC sr (4.89)
Tsrrs CCLCLC 12
12
(4.90)
3
33
33
3
3
13
000000
05.200000
005.20000
000000
000000
000000
000000
3
3
ls
ls
ls
ls
ls
ls
ls
s
L
ML
ML
L
L
L
L
CLC (4.91)
3
33
33
3
3
3
3
13
000000
05.200000
005.20000
000000
000000
000000
000000
lr
lr
lr
lr
lr
lr
lr
r
L
ML
ML
L
L
L
L
CLC (4.92)
0000000
00000
00000
0000000
0000000
0000000
0000000
5.2
33
33
31
3
CosSin
SinCos
MCLC sr (4.93)
Tsrrs CLCCLC 13
13
(4.94)
88
On the basis of above equations, the developed form of stator and rotor voltage equation can
be written as:
1 2 3( )MC
MC MCs s s
dV R R R i
dt
(4.95)
)(
)()(
1111111
11
1111321321
rrmr
m
rmmisisis
MCsss
MC
iCosiSinLdt
diSinL
dt
diCosL
dt
diLLLLiRRRV
(4.96)
dt
diRRRV
MCMC
sssMC
)( 321
(4.97)
)(
)()(
1111111
11
1111321321
rrmr
m
rm
MC
misisisMC
sssMC
iSiniCosLdt
diCosL
dt
diSinL
dt
diLLLLiRRRV
(4.98)
dt
diRRRV
MCxMC
xsssMCx
113211 )(
(4.99)
)(
)()(
2222222
12
2122
1232113211
rrmr
m
rxm
MCx
misisisMCxsss
MCx
iCosiSinLdt
diSinL
dt
diCosL
dt
diLLLLiRRRV
(4.100)
dt
diRRRV
MCyMC
ysssMCy
113211 )(
(4.101)
)(
)()(
2222222
22
1122
1232113211
rrmr
m
rym
MCy
misisisMCysss
MCy
iSiniCosLdt
diCosL
dt
diSinL
dt
diLLLLiRRRV
(4.102)
dt
diRRRV
MCxinv
xsssMCx
223212 )(
(4.103)
89
)(
)()(
3333333
33
3233
2332123212
rrmr
m
rxm
MCx
misisisMCxsss
MCx
iCosiSinLdt
diSinL
dt
diCosL
dt
diLLLLiRRRV
(4.104)
dt
diRRRV
MCyinv
ysssMCy
223212 )(
(4.105)
3333333
33
1233
2332123212
(
)()(
rrmr
m
rym
MCy
misisisMCysss
MCy
iSiniCosLdt
diCosL
dt
diSinL
dt
diLLLLiRRRV
(4.106)
Rotor Equations of 1st machine
)(
)(0
1111
1111111111
MCMCm
drmr
MC
m
MC
mrrr
iCosiSinL
dt
diLL
dt
diSinL
dt
diCosLiRV
(4.107)
)(
)(0
1111
1111111111
MCMCm
rmr
MC
m
MC
mrrr
iSiniCosL
dt
diLL
dt
diCosL
dt
diSinLiRV
(4.108)
dt
diLiRV
MCrx
irrxrrx11
11111 10
(4.109)
dt
diLiRV
MCry
irryrry11
11111 10
(4.110)
dt
diLiRV
MCrx
irrxrrx12
11212 10
(4.111)
dt
diLiRV
MCry
irryrry12
11212 10
(4.112)
90
dt
diLiRV
MCr
irrrr1
111
0010 0
(4.113)
Rotor Voltage Equations of machine 2:
2 2 1 2 2 2 2
2 2 2 2 2 2 2
0 ( ) ( )
( ) ( ) ( )
MC MCr r r m m
MC MCr m r m
V R i L Cos pi L Sin pi
Li L pi L Sin i Cos i
(4.114)
2 2 2 2 2 2 2
2 2 2 2 2 2 2
0 ( ) ( )
( ) ( ) ( )
MC MCr r r m m
MC MCr m r m
V R i L Sin pi L Cos pi
Li L p L Sin i Sin i
(4.115)
1 2 1 1 22 2 20 MCx r r x r ir x rV R i L pi
(4.116)
1 1 1 22 2 2 20 MCy r r y r ir y rV R i L pi
(4.117)
2 2 2 2 2 22 20 MCx r r x r ir x rV R i L pi
(4.118)
2 2 2 2 2 22 20 MCy r r y r ir y rV R i L pi
(4.119)
2 2 20 2 2 0 00 MCr r r ir rV R i L pi
(4.120)
Rotor Voltage Equation of Machine 3:
3 3 3 3 3 3 3 3 3 3
3 3 3 3
0 ( ) ( ) ( )
( ) ( )
MC MCr r r m m r m r
MC MCm
V R i L Cos pi L Sin pi Li L pi
L Sin i Cos i
(4.121)
3 3 3 3 3 3 3
3 3 3 3 3 3 3
0 ( ) ( )
( ) ( ) ( )
MC MCr r r m m
MC MCr m r m
V R i L Sin pi L Cos pi
Li L pi L Sin i Sin i
(4.122)
1 1 13 3 3 3 30 MCx r r x r ir x rV R i L pi
(4.123)
1 1 1 33 3 3 30 MCy r r y r ir y rV R i L pi
(4.124)
2 2 23 3 3 3 30 invx r r x r ir x rV R i L pi
(4.125)
91
2 2 3 23 3 3 30 MCy r r y r ir y rV R i L pi
(4.126)
3 30 3 3 0 0 30 MCr r r ir rV R i L pi
(4.127)
The motor electromagnetic torque is entirely developed due to the interaction of d-q current
components and is independent of x-y current components. Since rotor is short circuited, the
x-y and zero sequence components of rotors are zero. Due to the star connection in stator
make stator zero sequence components to zero. The motor torque equation can be written as:
)())((
)())((
)())((
333333333
222222222
111111111
MCr
MCr
MCr
MCrme
MCr
MCr
MCr
MCrme
MCr
MCr
MCr
MCrme
iiiiSiniiiiCosLPT
iiiiSiniiiiCosLPT
iiiiSiniiiiCosLPT
(4.128)
4.4 c Transformation of model in to the stationary common reference frame
Stator and rotor transformation matrices are 7x7 matrices but, only d-q components need to
transform to the stationary common reference frame as other x-y and zero sequence
components are zero. The rotational transformation matrices are given as in equation (4.23);
1000000
0100000
0010000
0001000
0000100
00000
00000
1
11
11
CosSin
SinCos
D
(4.129)
1000000
0100000
0010000
0001000
0000100
00000
00000
2
22
22
CosSin
SinCos
D (4.130)
92
1000000
0100000
0010000
0001000
0000100
00000
00000
3
33
33
CosSin
SInCos
D
(4.131)
Here, Angle of transformation
1 1 1dt , 2 2 2dt and 3 3 3dt (4.132)
So the new variables are defined as:
1 11
r rdqV D V , 2 2
2r rdqV D V , 3 3
3r rdqV D V
(4.133)
1 11
r rdqi D i , 2 2
2r rdqi D i , 3 3
3r rdqi D i
(4.134)
1 1 1 1 1 11 1 1
1 2 1 2 1 22 1 1
1 3 1 3 1 33 1 1
MC MC MC
r r rdq dq dq
r r rdq dq dq
r r rdq dq dq
V i i
D V D i D idV R L
dtD V D i D i
D V D V D V
(4.135)
Here, Angles of transformation 1 1 1dt and 2 2 2dt
So the new variables are defined as 1 11
r rdqV D V , 2 2
2r rdqV D V , 3 3
3r rdqV D V
(4.136)
1 11
r rdqi D i , 2 2
2r rdqi D i , 3 3
3r rdqi D i
(4.137)
1
2
3
0 0 0
0 0 0
0 0 0
0 0 0
I
DD
D
D
(4.138)
Pre-multiplying both sides by ‘D’ one obtain:
( )dq dqdq dq dq
d L IV Ri Gi
dt
(4.139)
where
93
0
0
0
MC
dq
V
V
(4.140)
1
2
3
MC
r
dq r
r
i
ii
i
i
(4.141)
1 2 3
1
2
3
0 0 0
0 0 0
0 0 0
0 0 0
s s s
r
r
r
R R R
RR
R
R
(4.142)
Individual matrices are given as:
1
1
1
11
1
1
1
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
s
s
s
ss
s
s
s
R
R
R
RR
R
R
R
(4.143)
1 2 3 1 2 3
1 1
2 2
3 3
0 0
0 0
0 0
s s s sr sr srdq dq dq dq dq dq
rs rdq dq
dq rs rdq dq
rs rdq dq
L L L L L L
L LL
L L
L L
(4.144)
Torque Equations
Torque equation of the machine become
1 1 1 1 1
2 2 2 2 1 2 1
3 2 2 3 2 3 2
( )
( )
( )
MC MCe m dr qr
MC MCe m dr y qr x
MC MCe m dr y qr x
T PL i i i i
T P L i i i i
T P L i i i i
(4.145)
94
By omitting zero sequence and x1-y1 and x2-y2 components from rotors and zero sequence
from power source, the complete model in the stationary common reference frame for the
three seven-phase series-connected induction machines is given as:
1 1 1 1 1 2 2 3 3( )MC MC MC MC MC MC MCs ls m m dr s ls s lsV R i L L pi L pi R i L pi R i L pi
1 1 1 1 1 2 2 3 3( )MC MC MC MC MC MC MCs ls m m qr s ls s lsV R i L L pi L pi R i L pi R i L pi
1 1 1 1 1 2 1 2 2 1 2 2 3 1 3 3 1( ) ( )MC MC MC MC MC MC MC MCx s x ls x s x ls m x m dr s x ls m xV R i L pi R i L L pi L pi R i L L pi (4.146)
1 1 1 1 1 2 1 2 2 1 2 2 3 1 3 3 1( ) ( )MC MC MC MC MC MC MC MCy s y ls y s y ls m y m qr s y ls m yV R i L pi R i L L pi L pi R i L L pi
2 1 2 1 2 2 2 2 2 2 2 3 3 2 3 3 2( ) ( )MC MC MC MC MC MC MC MCx s x ls x s x ls m x m dr s x ls m xV R i L pi R i L L pi L pi R i L L pi
2 1 2 1 2 2 2 2 2 2 2 3 3 2 3 3 2( ) ( )MC MC MC MC MC MC MC MCy s y ls y s y ls m y m qr s y ls m yV R i L pi R i L L pi L pi R i L L pi
Rotor Equations of machine 1 is obtained as:
1 1 1 1 1 1
1 1 1 1 1
0 ( )
( ( ) )
MCr dr m lr m dr
MCm lr m qr
R i L pi L L pi
L i L L i
(4.147)
1 1 1 1 1 1
1 1 1 1 1
0 ( )
( ( )
MCr qr m r m qr
MCm lr m dr
R i L pi L L pi
L i L L i
(4.148)
Rotor Equations of machine 2 is obtained as:
2 2 2 2 2 2
2 2 2 2 2
0 ( )
( ( )
MCr dr m r m dr
MCm lr m qr
R i L pi L L pi
L i L L i
(4.149)
1 2 2 2 2 2
2 2 2 2 2
0 ( )
( ( )
MCr qr m r m qr
MCm lr m dr
R i L pi L L pi
L i L L i
(4.150)
95
Rotor Equations of machine 3 is given as:
3 3 3 3 3 3
3 3 3 3 3
0 ( )
( ( ) )
MCr dr m r m dr
MCm lr m qr
R i L pi L L pi
L i L L i
(4.151)
3 3 3 3 3 3
3 3 3 3 3
0 ( )
( ( ) )
MCr qr m r m qr
MCm lr m dr
R i L pi L L pi
L i L L i
(4.152)
It is observed from the rotor equations of all three machines that the interaction of rotors and
stator current is only in the α-β components. There is no interaction of x1-y1 or x2-y2
components of stator and rotor. The complete seven-phase three-motor drive system can thus
be represented by equations (4.146) to (4.152). Hence, these voltage equations along with the
torque equations of (4.153) and the following electromechanical equations can completely
describe the system of Fig. 4.5
dt
d
P
JTT Le
1
1
111
dt
d
P
JTT Le
2
2
222
(4.153)
dt
d
P
JTT Le
3
3
333
4.4d Simulation Result
To validate the mathematical model developed in the previous section, equations (4.146) to
(4.153) are simulated in MATLAB/SIMULINK. Since the aim is to validate the mathematical
model, ideal seven-phase sinusoidal source is considered for simulation purpose. The source
voltage is assumed as 220 V rms per phase, 50 Hz. The three machines are chosen identical
with stator resistance of 10 Ω, rotor resistance of 6.3 Ω, leakage inductance of 40 mH,
magnetizing inductance of 420 mH, inertia 0.01 and number of poles 4, and rated torque
equal 11.662 Nm. Since the three machines are decoupled from each other, they can operate
in different conditions. Three machines are simulated under v/f = constant control conditions.
Machine-1 is allowed to run at rated speed (50 Hz) with applied voltage equal to rated value
of 220 V rms. Machine-2 is allowed to run at half of the rated speed (25 Hz) with applied
voltage of 110 V rms. Machine-3 is run at slightly less than quarter of the rated speed (12 Hz)
and the applied voltage is accordingly reduced. The resulting waveforms are presented in
Figs. 4.6a-c. The applied voltage of phase ‘a’ is presented in Fig.4.13 It comprises of three
different frequencies voltages corresponding to the operating conditions of the three
96
machines. The transformed source currents (assumed ideal sinusoidal currents) are depicted
in Figs. 4.10-4.12, the three components are shown. It is seen from Figs. 4.10-4.12, that the
frequencies and magnitude of each pair of currents are different and are matching the set
operating conditions of the three machines. Fig. 4.6 shows the speed and torque responses of
the three series connected machines. The response shows typical v/f = constant control
behavior. The machine accelerates to the set speeds and the torque response is typical.
Fig. 4.13 illustrate the spectrum of the source side voltage of phase ‘a’ and its transformed
components. The phase ‘a’ voltage shows three fundamental components at three different
frequencies with different magnitudes. The frequency components match the commanded
values. The transformed components show just one fundamental component. This clearly
indicates the independence of control of three machines. The phase ‘a’ voltage components
appear in three different planes that will control each machine individually. This validates the
mathematical model developed in the previous section. The independent control of each
machine is seen from the simulation results. The results also validate the model of derived in
the previous section.
a.
b.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-10
-5
0
5
10
15
Time (s)
Tor
que
M1
(Nm
)
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-10
-5
0
5
10
15
Time (s)
Tor
que
M2
(Nm
)
97
c.
d.
Fig. 4.6. Response of Seven-phase Three-motor drive supplied by ideal voltage source.
he configuration of seven phase 3 motor system is verified by SIMULINK modeling. Result
is verified with loading and reversing effect, shown in Fig (4.7)-(4.16).
Fig. 4.7, Torque and speed characteristics of Machine 1.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-4
-2
0
2
4
6
Time (s)
Tor
que
M3
(Nm
)
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-200
0
200
400
600
800
1000
1200
1400
1600
Time (s)
Spe
ed M
1,M
2 an
d M
3 (
rpm
)
Speed 1
IM1
IM2
IM3
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8-2000
0
2000
Spe
ed M
1 (r
pm)
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8-50
0
50
Time (s)
Tor
que
M1
(Nm
)
Forward LoadingReverse
98
Fig. 4.8. Torque and Speed characteristics of Machine 2.
Fig. 4.9. Speed and Torque Characteristics of Machine 3.
Fig. 4.10. Current ‘Iα-Iβ’.
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8-40
-20
0
20
Time (s)
Tor
que
M2
(Nm
)
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8-1000
0
1000
Spp
ed M
2 (r
pm)
Forward LoadingReverse
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-500
0
500
Spe
ed M
3 (r
pm)
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-10
-5
0
5
10
Time (s)
Tor
que
M3
(Nm
)
0 0.1 0.2 0.3 0.4 0.5 0.6-10
-5
0
5
10
I (
A)
0 0.1 0.2 0.3 0.4 0.5 0.6-10
-5
0
5
10
Time (s)
I (
A)
99
Fig. 4.11. Current ‘Ix1-Iy1’.
Fig. 4.12. Source current ‘Ix2-Iy2’.
Fig. 4.13. Spectrum of source phase ‘a’ voltage.
0 0.1 0.2 0.3 0.4 0.5 0.6-10
-5
0
5
10
I y1 (
A)
0 0.1 0.2 0.3 0.4 0.5 0.6-10
-5
0
5
10
Time (s)
I x1 (
A)
0 0.1 0.2 0.3 0.4 0.5 0.6-4
-2
0
2
4
I x2 (
A)
0 0.1 0.2 0.3 0.4 0.5 0.6-4
-2
0
2
4
Time (s)
I y2 (
A)
100
Fig. 4.14. Spectrum of source voltage Vα.
Fig. 4.15. Spectrum of source voltage Vx1.
Fig. 4.16. Spectrum of source voltage Vx2.
0.54 0.55 0.56 0.57 0.58 0.59-1000
-500
0
500
1000
V [
V]
Time (s)
0 100 200 300 400 500 600 700 800 900 10000
200
400
600
Fundamental = 582.0653
FF
T (
V)
Frequency (Hz)
0.84 0.86 0.88 0.9 0.92 0.94 0.96 0.98 1 1.02-400
-200
0
200
400
Vx1
[V
]
Time (s)
0 20 40 60 80 100 120 140 160 180 2000
100
200
300
Fundamental = 291.0326
FF
T V
x1
Frequency (Hz)
0.65 0.7 0.75 0.8 0.85 0.9 0.95 1-200
-100
0
100
200
Vx2
[v]
Time (s)
0 20 40 60 80 100 120 140 160 180 2000
50
100
150
Fundamental = 132.2882
FF
T V
x2
Frequency (Hz)
101
It is seen from above figures that the phase voltage (Matrix Converter output voltage) has
three different fundamental frequency components at 12 Hz, 25 Hz and 50 Hz corresponding
to the operating speeds of the three series-connected induction machines. These fundamental
components appear as d-q, x1-y1 and x2-y2 components and there is no interaction between
them. This proves the correctness of the developed mathematical model and the concept of
independence of control.
4.5 Summary
This chapter developed and reported complete mathematical models of multi-phase multi-
motor drive system. Three different configurations are considered namely five-phase, six-
phase and seven-phase systems. Phase variable models are developed and then transformed to
appropriate number of axes. This is done in order to eliminate the position dependence of
inductances. The developed mathematical models showed that the torque can be produced by
using only two orthogonal components of currents. Hence in case of a five-phase and six-
phase systems one pair of currents are free to use (since there are two sets of currents) in
controlling the second machine. Thus stator windings of two-motors are connected in series
or in parallel and are supplied by only one power converter and both motors are controlled
independently. In similar fashion, a seven-phase system is seen to control three series or
parallel connected machines. The developed mathematical models are verified using
simulation approach.
References
[4.1] M. Jones, S.N. Vukosavic, E. Levi, A. Iqbal, “A six-phase series connected two-motor drive with decoupled dynamic control,” IEEE Trans. Ind. Appl., vol. 41, no. 4, Jul./Aug. 2005a, pp. 1056–1066.
[4.2] M. Jones, E. Levi, “Series connected quasi six-phase two-motor drives with independent control”, Math Comput. Simul. (Trans. IMACS), vol. 71, no. 4-6, June 2006, pp. 415-424.
[4.3] A. Iqbal, “Modeling and Control of Five-phase and six-phase series-connected two-motor drive system”, PhD Thesis, Liverpool John Moores University, Liverpool, UK, 2006.
[4.4] E. Levi, M. Jones, S.N. Vukosavic, A. Iqbal, H.A. Toliyat, “Modelling Control and Experimental investigation of a five-phase series-connected two-motor drive with
102
single inverter supply,” IEEE Trans. On Industrial Electronics, vol. 54, no. 3, , pp. 1504-1516, June 2007.
[4.5] E. Levi, M. Jones, A. Iqbal, H.A. Toliyat, “An induction machine/syn-rel two-motor five-phase series-connected drive,” IEEE Trans. On Energy Conversion, vol. 22, no. 2, June 2007, 281-289.
[4.6] M. Jones, “A novel concept of series connected multi-phase, multi-motor drive systems”, PhD. thesis, School of Engineering, Liverpool John Moores University, Liverpool, UK in Jan 2005.
[4.7] J.W.Kelly, E.G.Strangas, J.M.Miller, “Multi-phase inverter analysis,” Proc. IEEE Int. Electric Machines and Drives Conf. IEMDC, Cambridge, MA, 2001, pp. 147-155.
[4.8] O. Ojo, G. Dong, “Generalized discontinuous carrier-based PWM modulation scheme for multi-phase converter-machine systems,” Proc. IEEE Ind. Appl. Soc. Annual Meeting IAS, Hong Kong, 2005, CD-ROM paper 38P3.
[4.9] A. Iqbal, E. Levi, M. Jones, S.N. Vukosavic, “Generalised sinusoidal PWM with harmonic injection for multi-phase VSIs,” Proc. IEEE Power Elec. Spec. Conf. PESC, Jeju, Korea, 2006, pp. 2871-2877.
[4.10] M.J. Duran, E. Levi, “Multi-dimensional approach to multi-phase space vector pulse width modulation,” Proc. IEEE Ind. Elec. Soc. Annual Meeting IECON, Paris, France, 2006, pp. 2103-2108.
[4.11] A. Iqbal, E. Levi, “Space vector PWM techniques for sinusoidal output voltage generation with a five-phase voltage source inverter,” Electric Power Components and Systems, vol. 34, no. 2, 2006, pp. 119-140.
[4.12] J.W. Kelly, E.G. Strangas, J.M. Miller, “Multi-phase space vector pulse width modulation,” IEEE Trans. on Energy Conversion, vol. 18, no. 2, 2003, pp. 259-264.
[4.13] H.M. Ryu, J.H. Kim, S.K. Sul, “Analysis of multi-phase space vector pulse-width modulation based on multiple d-q spaces concept,” IEEE Trans. on Power Electronics, vol. 20, no. 6, 2005, pp. 1364-1371.
[4.14] D. Casadei, G. Serra, A. Tani, L. Zarri, “Multi-phase inverter modulation strategies based on duty-cycle space vector approach,” Proc. Ship Propulsion and Railway Traction Systems Conf. SPRTS, Bologna, Italy, 2005, pp. 222-229.
[4.15] P.S.N. De Silva, J.E. Fletcher, B.W. Williams, “Development of space vector modulation strategies for five-phase voltage source inverters,” Proc. IEE Power Electronics, Machines and Drives Conf. PEMD, Edinburgh, UK, 2004, pp. 650-655.
[4.16] A. Iqbal, E. Levi, “Space vector modulation schemes for a five-phase voltage source inverter,” Proc. European Power Electronics and Applications Conf. EPE, Dresden, Germany, 2005, CD-ROM paper 0006.
[4.17] X. Kestelyn, E. Semail, J.P. Hautier, “Multi-phase system supplied by SVM VSI: A new fast algorithm to compute duty cycles,” EPE Journal, vol. 14, no. 3, 2004, pp. 25-31.
[4.18] P. Delarue, A. Bouscayrol, E. Semail, “Generic control method of multi-leg voltage-source converters for fast practical implementation,” IEEE Trans. on Power Electronics, vol. 18, no. 2, 2003, pp. 517-526.
[4.19] S. Xue, X. Wen, Z. Feng, “A novel multi-dimensional SVPWM strategy of multi-phase motor drives,” Proc. Power Electronics and Motion Control Conf. EPE-PEMC, Portoroz, Slovenia, 2006, pp. 931-935.
[4.20] O. Ojo, G. Dong, Z. Wu, “Pulse-width modulation for five-phase converters based on device turn-on times,” Proc. IEEE Ind. Appl. Soc. Annual Meeting IAS, Tampa, FL, 2006, CD-ROM paper IAS15p7.
103
[4.21] S. Xue, X. Wen, “Simulation analysis of two novel multi-phase SVPWM strategies,” Proc. IEEE Int. Conf. on Industrial Technology ICIT, Hong Kong, 2005, pp. 1401-1406.
[4.22] D. Casadei, D.Dujić, E.Levi, G.Serra, A.Tani, L.Zarri, “General modulation strategy for seven-phase inverters with independent control of multiple voltage space vectors," IEEE Trans. on Industrial Electronics, vol. 55, no. 5, 2008, pp. 1921-1932.
[4.23] D. Dujic, G. Grandi, M. Jones, E. Levi; “A space vector PWM scheme for multi-frequency output voltage generation with multi-phase voltage source inverters,” IEEE Trans. on Industrial Electronics, vol. 55, no. 5, 2008, pp. 1943-1955.
[4.24] M. Duran, S. Toral, F. Barrero, E. Levi, “Real-time implementation of multi-dimensional five-phase space vector pulse-width modulation,” Electronics Letters, vol. 43, no. 17, 16th August 2007, pp. 949-950.
[4.25] E. Levi, D. Dujic, M. Jones, G. Grandi, “Analytical determination of DC-bus utilization limits in multi-phase VSI supplied AC drives,” IEEE Trans. on Energy Conversion, vol. 23, no. 2, 2008, pp. 433-443.
[4.26] M. Jones, S.N. Vukosavic, E. Levi, “Parallel-connected multi-phase multi-drive systems with single inverter supply,” IEEE Transactions on Industrial Electronics, vol. 56, no. 6, June 2009, pp. 2047-2057.
[4.27] M. Jones, E. Levi, S.N. Vukosavic, H. A. Toliyat, “Independent vector control of a seven-phase three-motor drive system supplied from a single voltage source inverter,” 34th Annual IEEE Power Electronics Specialists Conference. ,vol. 4, Jun 2003, pp. 1865-1870.
[4.28] S. Moinuddin, A. Iqbal, M.R. Khan, “Space vector approach to model a seven-phase voltage source inverter,” Proceedings of IEEE International Conference on Recent Advancements and Applications of Computer in Electrical Engineering (RACE 2007) Bikaner, India , 2007, CD-ROM. Paper F-398.
[4.29] S. Moinuddin, A. Iqbal, "Modelling and Simulation of a Seven-Phase Voltage Source Inverter Using Space Vector Theory,” I-manager's Journal on Electrical Engineering, vol. 1, No. 1, July – September 2007, pp 30 – 41.
[4.30] D. Casadei, D. Dujic E. Levi G. Serra, A. Tani, L. Zarri, “general modulation strategy for seven phase inverters with independent control of multiple voltage space vectors”, IEEE Trans. Ind. Elect., vol. 55, no. 5, May 2008, pp. 1921-1932.
104
Chapter 5 Space Vector Modelling of Multi-phase
Matrix Converter
5.1 Introduction
This chapter discusses the modelling procedures for a Matrix Converter with three-phase
input and five-phase, six-phase and seven-phase outputs. The model is developed in space
vector form. At first the topology of three-phase input and five-phase output is considered.
The switching combinations are identified. The total number of possible switching
combinations is obtained as 2n where n is the number of power semiconductor switching
devices being used. However, all these switching states are not permitted due to the fact that
the input source should not be short-circuited and the output phases should not be open-
circuited. The load is considered to be inductive and hence this safety condition is imposed.
With such constraints, the number of permitted switching states reduces. Further, the space
voltage vectors that arise due to these switching combinations are analyzed on the basis of
their amplitude and frequency. They are grouped according to the number of output and input
phase connections. It is observed that amplitude and frequency are variable in several cases.
It is concluded that the space voltage vectors whose amplitude and frequency are constant,
are only viable for space vector PWM implementation. The other two topologies analyzed are
three-phase input and six-phase output, and three-phase input and seven-phase output.
5.2 Space Vector Model of a Three-phase to Five-phase Matrix Converter
The space vector model of a multi-phase Matrix Converter (three-phase input and five-phase
output) is developed on the basis of representation of the three-phase input current and five-
phase output line voltages on the space vector plane. The input current is three-phase and
hence the traditional space vector model is sufficient to describe the input side. The output is
five-phase and hence five-phase output voltage is to be modelled in five dimensional space.
During Matrix Converter operation all output phases are connected to each input phases
depending on the state of the switches. For a three-phase input and five-phase output Matrix
Converter, total numbers of bi-directional power semiconductor switches (IGBTs) are fifteen
(three switches per leg). One input line is connected to five bi-directional switches and hence
three input line is connected to fifteen switches in all. A small LC filter is connected at the
105
source side (three-phase side) to remove the switching ripples from the source current
waveforms, thus making it sinusoidal. A general switching circuit is shown in Fig. 5.1. The
filter is not shown in the Fig. 5.1. Each switch is shown with two sided arrows in order to
show its bi-directional nature. There are several methods to realize the bi-directional power
switches as discussed in reference [5.1] - [5.3].
AV FVDVCVBV
aV bV cV
Fig. 5.1. Basic topology of a three-phase to five-phase Matrix Converter.
With this number of switches a total combination of switching can be made in the range of
152 = 32768. However, out of these many possible switching states, not all are permissible.
To obtain the permissible switching states, one has to consider the safety of the source side as
well as load side. It is well known that the switching states that cause a short circuit at the
source side are not allowed. Similarly, opening of the load side is also not permitted, since
the loads are mostly inductive and sudden opening of inductive circuit causes high voltage
spikes that may damage the switching devices. . Hence, for the safe operation of Matrix
Converter protecting the source and load, the following conditions must be considered at any
switching time:
Input phases (three-phase side) should never be short circuited,
Output phases (five-phase side) should never be open circuited.
106
Considering the above two major constraints, there are 53 = 243, different switching
combinations for connecting output phases to input phases. It is important to note here that
the number of switching combinations is same as that of a three-level five-phase inverters as
given in reference [5.4]. Therefore, one can say that the operation of a Matrix Converter is
similar to the operation of a three-level back-to-back converter. These switching
combinations can be put in five different groups or in other words, all these vectors are
divided into five different groups. The space vector group nomenclature is defined as [x,y,z],
where x represent the number of output phase connected to input phase ‘a’, y represent the
number of output phase connected to input phase ‘b’ and z shows the number of output
phases being connected to input phase ‘c’.
1. [5, 0, 0]: All the five output phases are connected to the same input phase. Since there
are three phases at the input side, so three possible situation are possible, i.e. all five
phases are connected to phase ‘a’ of input, or phase ‘b’ of input or finally to phase ‘c’
of input. Hence, this group consists of three different switching combinations. These
vectors have zero magnitude and zero frequency, therefore they are termed as zero
vectors.
2. [4, 1, 0]: Four of the output phases are connected to the same input phase and the fifth
output phase is connected to any of the other two remaining input phases, e.g. if
ABCD is connected to phase ‘a’ then ‘e’ can either connect with phase ‘b’ or phase
‘c’. This group consists of total 30 switching combinations. The space vectors
produced due to these switching combinations have variable amplitude at a constant
frequency in space. It means amplitude of the output voltages depend on the selected
input line voltages. In this case the phase angle of the output voltage space vector
does not depend on the phase angle of the input voltage space vector. The 30
combinations in this group determine 10 pre-fixed positions of the output voltage
space vectors which are not dependent on i (input voltage vector position as shown
in Fig.5.5). Similar condition is also valid for current vectors. The 30 combinations in
this group determine 6 prefixed positions of the input current space vectors which are
not dependent on o (output current vector position). The input side space vectors are
not shown.
3. [3, 2, 0]: Three of the output phases are connected to the same input phase and the
two other output phases are connected to any one of the other two input phases. This
group consists of 60 switching combinations. These vectors have also variable
107
amplitude at a constant frequency in space. It means amplitude of the output voltages
depend on the selected input line voltages. Also in this case the phase angle of the
output voltage space vector does not depend on the phase angle of the input voltage
space vector. The 60 combinations in this group determine 10 prefixed positions of
the output voltage space vectors which are not dependent on i . Similar condition is
also valid for current vectors. The 60 combinations in this group determine 6 prefixed
positions of the input current space vectors which are not dependent on o .
4. [3, 1, 1]: Three of the output phases are connected to the same input phase and the
two other output phases are connected to the other two input phases respectively. This
group consists of total 60 switching combinations. These vectors have variable
amplitude variable frequency in space. It means amplitude of the output voltages
depend on the selected input line voltages. In this case the phase angle of the output
voltage space vector depends on the phase angle of the input voltage space vector.
The 60 combinations in this group do not determine any prefixed positions of the
output voltage space vector. The locus of the output voltage space vectors form
ellipses in different orientations in space as i is varied. Similar condition is also
valid for current vectors. For the space vector modulation technique these switching
states are not used in the Matrix Converter since the phase angle of both input and
output vectors cannot be controlled independently. This control condition is important
in Matrix Converter operation.
5. [2, 2, 1]: Two of the output phases are connected to the same input phase and the two
other output phases are connected to another input phase and the fifth output phase is
connected the remaining third input phase. This group consists of altogether 90
switching combinations. The space vectors generated due to these switching
combinations produces also variable amplitude variable frequency in space. That is
the amplitude of the output voltages depend on the selected input line voltages. In this
case the phase angle of the output voltage space vector depends on the phase angle of
the input voltage space vector. The 90 combinations in this group do not determine
any prefixed positions of the output voltage space vector. The locus of the output
voltage space vectors form ellipses in different orientations in space as i is varied
i.e. with the rotation of the input space vector. Similar condition is also valid for
current vectors. For the space vector modulation technique these switching states are
108
also not used in the Matrix Converter since the phase angle of both input and output
vectors cannot be controlled independently.
The active switching states that produce constant amplitude voltage and constant frequency
that can be further used for space vector pulse width modulation schemes are:
Group 1 :[5, 0, 0] consists of 3 space voltage vectors,-Zero vectors
Group 2 :[4, 1, 0] consists of 30 space voltage vectors,-Active vectors
Group 3 :[3, 2, 0] consists of 60 space voltage vectors,-Active vectors.
As such total 93 space voltage vector combinations can be used to realize the space vector
modulation technique.
The switching circuit condition conditions corresponding to Group 1 ([5,0,0]), is illustrated in
Fig. 5.2. The bold circle indicates the connection of the input and output phases.
Fig. 5.2. Switching states of Group-1.
The switching circuit condition conditions corresponding to Group 2 ([4,1,0]), is illustrated in
Fig. 5.3. The bold circle indicates the connection of the input and output phases.
109
110
b
A
B
C
D
E
ca
abaaa
b
A
B
C
D
E
ca
bcbbb
b
A
B
C
D
E
ca
cacccb
A
B
C
D
E
ca
aabaa
b
A
B
C
D
E
ca
bbcbb
b
A
B
C
D
E
ca
ccacc
b
A
B
C
D
E
ca
cccac
b
A
B
C
D
E
ca
bbbcb
b
A
B
C
D
E
ca
aaaba
Fig. 5.3. Switching states of Group-2 [4,1,0].
The switching circuit condition conditions corresponding to Group 3 ([3,2,0]), is illustrated in
Fig. 5.4. The bold circle indicates the connection of the input and output phases.
111
112
113
114
115
116
Fig. 5.4. Switching states for Group-4 [3,2,0].
117
The permitted voltage space vectors are then transformed into two set of orthogonal planes
namely α-β and x-y in the stationary common reference frame. The transformation matrix
used as given in reference [5.4]:
2
1
2
1
2
1
2
1
2
12sin4sin4sin2sin0
2cos4cos4cos2cos1
sin2sin2sinsin0
cos2cos2coscos1
5
2
sA (5.1)
The 93 space voltage vectors that are finally used to implement space vector pulse width
modulation scheme are given in Fig. 5.5 in the orthogonal planes.
118
Fig. 5.5. Output voltage space vectors corresponding to the permitted switching combinations.
The large length space voltage vectors are shown in Fig. 5.6.
119
'oV
oV
"oV
Fig. 5.6. Output large length voltage space vectors.
The large length space voltage vectors detail are tabulated in Table 5.1. The table shows the
output phases connected to the input phase, the relation between the output line voltage and
input line voltage, the relation between the output line current and input line current, the
output voltage and the position. e.g. column 3, row 1, the input phase ‘a’ is connected to the
output phase ‘A’. The output line voltage, VAB is same as input line voltage Vab, the input
current Ia is same as output line current IA. The output voltage vector Vo is equal to
0.7608Vab (input line voltage), the position of vector is -18, and the input current vector is
1.156*iA and its position is -30 . It is to be noted that ‘l’ in Table 5.1 is equal to
7608.0
2
552
5
2
120
Table 5.1. Large length vectors.
121
The medium length space vectors are also analysed and the set is shown in Fig. 5.7
The medium length space voltage vectors details are tabulated in Table 5.2. The table shows
the output phases connected to the input phase, the relation between the output line voltages
and output line voltages, the relation between the output line current and input line current,
the output voltage and the position. e.g. column 3, row 1, the input phase ‘a’ is connected to
the output phase ‘A’. The output line voltage, VAB is same as input line voltage Vab, the input
current Ia is same as output line current IA. The output voltage vector Vo is 0.4706xVab
(input line voltage), the position of vector is 54, and the input current vector is 1.156*iA and
its position is -30. Note that m =
4702.02
552
5
2
.
122
Fig. 5.7. Output medium length voltage space vectors.
Table 5.2. Medium length vectors
123
124
The small vectors are also analysed and are shown in Fig. 5.8.
Fig. 5.8. Output small length voltage space vectors.
The small length space voltage vectors details are tabulated in Table 5.3. The table shows the output phases connected to the input phase, the relation between the output line voltages and output line voltages, the relation between the output line current and input line current, the output voltage and the position. e.g. column 3, row 1, the input phase ‘a’ is connected to the output phase ‘A’. The output line voltage, VAB is same as input line voltage Vab but in opposite phase, the input current Ia is same as output line current IA. The output voltage
125
vector Vo is 0.2906 x Vab (input line voltage), the position of vector is -90 , and the input current vector is 1.156*iA and its position is -30. Note that
where s = 2906.072Cos2.54Cos25
2 00 .
Table 5.3. Small space voltage vectors.
126
For each combination, the input and output line voltages can be expressed in terms of space
vectors as;
127
iji
j
ca
j
bcabi eVeVeVVV
...3
2 3
4
3
2
(5.2)
OjO
j
EA
j
DE
j
CD
j
BCABo eVeVeVeVeVVV
.....5
2 5
8
5
6
5
4
5
2
(5.3a)
OjO
j
EA
j
DE
j
CD
j
BCABoxy eVeVeVeVeVVV
.....5
2 5
16
5
12
5
8
5
4
(5.3b)
In the same way, the input and output line currents result is as follows:
iji
j
c
j
bai eIeIeIII
...3
2 3
4
3
2
(5.4)
OjO
j
E
j
D
j
C
j
BAo eIeIeIeIeIII
.....5
2 5
8
5
6
5
4
5
2
(5.5)
5.3 Space Vector Model of Three-phase to Six-phase Direct Matrix Converter
The space vector algorithm is based on the representation of the three phase input current and
five-phase output line voltages on the space vector plane. In Matrix Converters output phase
is connected to each input phase depending on the state of the switches. For a three to six-
phase Matrix Converter, total numbers of switches are eighteen. The circuit topology is
shown in Fig. 5.9.
Fig. 5.9. Three-six phase direct Matrix Converter.
128
With this number of switches a total combination of switching can be made in the range of
122 = 4096. For the safe switching in the Matrix Converter:
Input phases should never be short circuited,
Output phases should never be open circuited, at any switching time.
Considering the above two rules, there are 63 i.e. 729 different switching combinations for
connecting output phases to input phases. These switching combinations can be analysed in
five groups.
The switching combinations are represented as: There are six switches and each switch is free
to connect any of the three phases of input. In this way there are 3*6=18 possible
connections. Every switch can have two states on (1) or off (0). In this way there are 218
switching combinations. Considering the constraints “that any two input cannot be shorted
and output phase cannot be opened throughout the operation” switching combination reduces
to 36=729 only.
These 729 switching combinations are analysed in seven groups that shows how many output
phases are connected from input phase a, b, c. It is represented as [x, y, z] where x, y and z
are no of output phases connected to input phase a, b and c.
Group-1: [6,0,0]:- In this group all the output phases are connected to the same input phase.
Following it there are three possible combinations that are either all are connected to input
phase a or b or c. [6,0,0], [0,6,0],[0,0,6] there are 3 possible switching.
Group-2: [5,0,1]:- In this group 5 output phases are connected to the same input phase and
remaining phase is connected to any of the remaining input phases. Thus there are 6C5 (5 out
of 6 output phases are connected to same input) and 1C1 (remaining one output from any one
input) (6!/5!*1!)=6 combinations. In this way [5,0,1], [5, 1,0], [0,5,1], [1,5, 0], [0,1,5],
[1,0,5]. (6!/5!*1!)=6 combinations for every state combining all total 6*6=36 switching
combinations
Group-3: In this group four of the output phases are connected to one of the three input
phase simultaneously. Two remaining output phases are connected to one input phase and
one input phase not connected. All the possible sub groups are [4,0,2],[4,2,0],[0,4,2],[2,4,
0],[0,2,4] and [2,0,4]. Hence the following combinational logic can be used to obtain the
possible number of switching and hence number of space voltage vectors:
129
6C4 *2C2 =(6!/4!*2!) * (2!/2!*0!) =15*6=90.
Group-4: In this group three of the output phases are connected to one of the three input
phase simultaneously. Three remaining output phases are connected to one input phase and
one input phase is not connected. All the possible sub groups are [3,3,0], [3,0,3] and [0,3,3].
Hence the following combinational logic can be used to obtain the possible number of
switching and hence number of space voltage vectors: 6C3 *
3C3 =(6!/3!*3!) * (3!/1!*3!)=20*3=60.
Group-5: In this group four of the output phases are connected to one of the three input
phase simultaneously. One output phase are connected to one input phase and the last
remaining input phase is connected to one output phase. All the possible sub groups are
[4,1,1], [1,4,1] and [1,1,4]. Hence the following combinational logic can be used to obtain the
possible number of switching and hence number of space voltage vectors: 6C4 *
2C1*1C1 =(6!/4!*2!) * (2!/1!*1!)=15*2*3=90
Group-6. In this group three of the output phases are connected to one of the three input
phase simultaneously. Two of the remaining output phases are connected to one input phase
and one input phase is connected to one output phase. All the possible sub groups are [3,1,2],
[3,2,1], [1,3,2], [1,3,2], [2,1,3], and [1,2,3]. Hence, the following combinational logic can be
used to obtain the possible number of switching and hence number of space voltage vectors: 6C3 *
3C2 *1C1 =(6!/3!*3!) * (3!/2!*1!) =30*3 *6=360
Group-7: In this group two of the output phases are connected to one of the three input phase
simultaneously. Two of the output phases are connected to one input phase and one input
phase is connected to the remaining two output phase. The group can be mentioned as [2,2,2].
Hence the following combinational logic can be used to obtain the possible number of
switching and hence number of space voltage vectors: 6C2 *
4C2 *2C2 =(6!/2!*4!) * (4!/2!*2!) =15*6*1 =90.
Thus total switching vectors are given as: Group 1 : 6, 0, 0 consists of 3 vectors, Group 2 : 5, 1, 0 consists of 36 vectors,
130
Group 3 : 4, 2, 0 consists of 90 vectors. Group 4 : 3,3, 0 consists of 60 vectors. Group 5 : 4, 1, 1 consists of 90 vectors, Group 6 : 3, 2, 1 consists of 360 vectors. Group 7 : 2, 2, 2 consists of 90 vectors.
In this way there are total= 3+36+90+60+90+360+90=729 switching combinations in which
the space voltage vectors produced from switching of Groups-1,2, and 3 are having varying
amplitude and fixed frequency. Such switching combinations are 3+36+90+60=189 which is
used as active switching. Remaining are having varying amplitude and frequency both so
they are not helpful to find out the output voltage. These 189 switching states to find out
voltage is given in table 5.4. The Table 5.4 shows the output line voltage of the Matrix
Converter having lowest phase difference from reference one that is VAB.
The obtained output line voltages from switching combinations can be used to find out
voltages and currents in d-q and x-y planes by using decoupling transformation matrix as
given the equations (5.6) to (5.9:
)(6
2 5432
FAEFDECDBCABdqVaVaVaVaaVVV
(5.6)
)(6
2 108642
11 FAEFDECDBCAByxVaVaVaVaVaVV
(5.7)
)(6
2 1512963
00 FAEFDECDBCABVaVaVaVaVaVV
(5.8)
6/2 pijea (5.9)
Table 5.4. Three to Six phase output voltage and input current.
Vec A B C D E F VAB VBC VCD VDE VEF VFA Vdq α Vxy Β Ii αi
0 a a a a a a 0 0 0 0 0 0 0 0 0 0 0 0 0 b b b b b b 0 0 0 0 0 0 0 0 0 0 0 0 0 c c c c c c 0 0 0 0 0 0 0 0 0 0 0 0
1 a b a a b a 1 -1 0 1 -1 0 0 Vab 0 1.1546 -30 0 0 2 a a b a a b 0 1 -1 0 1 -1 0Vab 180 1.1546 90 0 0 3 c c b c c b 0 -1 1 0 -1 1 0 Vbc 0 1.1546 -90 0 90 4 c b c c b c -1 1 0 -1 1 0 0 Vbc -180 1.1546 150 0 90 5 c a c c a c 1 -1 0 1 -1 0 0 Vca 0 1.1546 -30 0 30 6 c c a c c a 0 1 -1 0 1 -1 0 Vca 180 1.1546 90 0 30 7 b b a b b a 0 -1 1 0 -1 1 0 Vab 0 1.1546 -90 0 -30 8 b a b b a b -1 1 0 -1 1 0 0 Vab -180 1.1546 150 0 -30 9 c b b c b b -1 0 1 -1 0 1 0 Vbc 0 1.1546 -150 0 -90
10 b c c b c c 1 0 -1 1 0 -1 0 Vbc 180 1.1546 30 0 90 11
a a c a
a c 0 -1 1 0 -1 1
0 Vca 0 1.1546 -90 0 -
150 12
c a a c a a 1 0 -1 1 0 -1
0 Vca 180 1.1546 30 0 -
150 13 b a a b a a -1 0 1 -1 0 1 0 Vab 0 1.1546 -150 0 150 14 a b b a b b 1 0 -1 1 0 -1 0Vab 180 1.1546 30 0 -30 15 b c b b c b 1 -1 0 1 -1 0 0 Vbc 0 1.1546 -30 0 -90
131
16 b b c b b c 0 1 -1 0 1 -1 0 Vbc 180 1.1546 90 0 -90 17 a c c a c c -1 0 1 -1 0 1 0 Vca 0 1.1546 -150 0 30 18
a c a a
c a -1 1 0 -1 1 0
0 Vca -180 1.1546 150 0 -
150 19 b a b a b a -1 1 -1 1 -1 1 0 Vab -11.3 0 -86.4 0 150 20 a b a b a b 1 -1 1 -1 1 -1 0Vab 168.7 0 93.58 0 -30 21 c b c b c b -1 1 -1 1 -1 1 0 Vbc -11.3 0 -86.4 0 -90 22 b c b c b c 1 -1 1 -1 1 -1 0 Vbc 168.7 0 93.58 0 90 23 a c a c a c -1 1 -1 1 -1 1 0 Vca -11.3 0 -86.4 0 30 24
c a c a c a 1 -1 1 -1 1 -1
0 Vca 168.7 0 93.58 0 -
150
25 a a b a a a 0 1 -1 0 0 0 0.33Vab 0 0.5773 90 1.155 150 26 a b a b a a 1 -1 1 -1 0 0 0.33Vab 0 0.5773 -90 1.155 150 27 a b b a b a 1 0 -1 1 -1 0 0.33Vab 0 0.9999 0 1.155 -30 28 a b b b a b 1 0 0 -1 1 -1 0.33Vab 0 0.5773 90 1.155 -30 29 a a a b a A 0 0 1 -1 0 0 0.33 Vab 60 0.5773 -150 1.155 150 30 a a b a b a 0 1 -1 1 -1 0 0.33 Vab 60 0.5773 30 1.155 150 31 a b a b b a 1 -1 1 0 -1 0 0.33 Vab 60 0.9999 -60 1.155 -30 32 a a b b a b 0 1 0 -1 1 -1 0.33Vab 60 0.9999 120 1.155 -30 33 a b b b b b 1 0 0 0 0 -1 0.33Vab 60 0.5773 30 1.155 -30 34 a a a a b A 0 0 0 1 -1 0 0.33Vab 120 0.5773 -30 1.155 150 35 a a b a b b 0 1 -1 1 0 -1 0.33Vab 120 0.9999 60 1.155 -30 36 a b a b b b 1 -1 1 0 0 -1 0.33Vab 120 0.5773 -30 1.155 -30 37 a a a b a b 0 0 1 -1 1 -1 0.33Vab 120 0.5773 150 1.155 150 38 a a a a a B 0 0 0 0 1 -1 0.33Vab 180 0.5773 90 1.155 150 39 a b a a b b 1 -1 0 1 0 -1 0.33Vab 180 0.9999 0 1.155 -30 40 a b a a a b 1 -1 0 0 1 -1 0.33Vab -120 0.5773 30 1.155 150 41 a b a a a a 1 -1 0 0 0 0 0.33Vab -60 0.5773 -30 1.155 150 42 a b b a a b 1 0 -1 0 1 -1 0.33Vab -60 0.9999 60 1.155 -30
43
a a a a a c 0 0 0 0 -1 1
0.33Vca 0 0.5773 -90 1.155 -
150 44
a a a a c a 0 0 0 -1 1 0
0.33Vca -60 0.5773 150 1.155 -
150 45
a a a c a a 0 0 -1 1 0 0
0.33Vca -120 0.5773 30 1.155 -
150 46
a a c a a a 0 -1 1 0 0 0
0.33Vca -180 0.5773 -90 1.155 -
150 47
a c a a a a -1 1 0 0 0 0
0.33Vca 120 0.5773 150 1.155 -
150 48
a a a c a c 0 0 -1 1 -1 1
0.33Vca -60 0.5773 -30 1.155 -
150 49
a c a
a a c -1 1 0 0 -1 1
0.33Vca 60 0.5773 -150 1.155 -
150 50
a a c a c a 0 -1 1 -1 1 0
0.33Vca -120 0.5773 -150 1.155 -
150 51
a c a c a a -1 1 -1 1 0 0
0.33Vca -180 0.5773 90 1.155 -
150 52 a a c a c c 0 -1 1 -1 0 1 0.33Vca -60 0.9999 -120 1.155 30 53 a c a a c c -1 1 0 -1 0 1 0.33Vca 0 0.9999 -180 1.155 30 54 a c a c c a -1 1 -1 0 1 0 0.33Vca -120 0.9999 120 1.155 30 55 a a c c a c 0 -1 0 1 -1 1 0.33Vca -120 0.9999 -60 1.155 30 56 a c c a a c -1 0 1 0 -1 1 0.33Vca 120 0.9999 -120 1.155 30 57 a c c a c a -1 0 1 -1 1 0 0.33Vca -180 0.9999 -180 1.155 30 58 a c a c c c -1 1 -1 0 0 1 0.33Vca -60 0.5773 150 1.155 30 59 a c c c a c -1 0 0 1 -1 1 0.33Vca -180 0.5773 -90 1.155 30 60 a c c c c c -1 0 0 0 0 1 0.33Vca -120 0.5773 -150 1.155 30
61 b b b b b c 0 0 0 0 1 -1 0.33Vbc 180 0.5773 90 1.155 -90 62 b b b b c b 0 0 0 1 -1 0 0.33Vbc 120 0.5773 -30 1.155 -90 63 b b b c b b 0 0 1 -1 0 0 0.33Vbc 60 0.5773 -150 1.155 -90 64 b b c b b b 0 1 -1 0 0 0 0.33Vbc 0 0.5773 90 1.155 -90 65 b c b b b b 1 -1 0 0 0 0 0.33Vbc -60 0.5773 -30 1.155 -90 66 b b b c b c 0 0 1 -1 1 -1 0.33Vbc 120 0.5773 150 1.155 -90 67 b c b b b c 1 -1 0 0 1 -1 0.33Vbc -120 0.5773 30 1.155 -90 68 b b c b c b 0 1 -1 1 -1 0 0.33Vbc 60 0.5773 30 1.155 -90 69 b c b c b b 1 -1 1 -1 0 0 0.33Vbc 0 0.5773 -90 1.155 -90 70 b b c b c c 0 1 -1 1 0 -1 0.33Vbc 120 0.9999 60 1.155 90 71 b c b b c c 1 -1 0 1 0 -1 0.33Vbc 180 0.9999 0 1.155 90 72 b c b c c b 1 -1 1 0 -1 0 0.33Vbc 60 0.9999 -60 1.155 90 73 b b c c b c 0 1 0 -1 1 -1 0.33Vbc 60 0.9999 120 1.155 90
132
74 b c c b b c 1 0 -1 0 1 -1 0.33Vbc -60 0.9999 60 1.155 90 75 b c c b c b 1 0 -1 1 -1 0 0.33Vbc 0 0.9999 0 1.155 90 76 b c b c c c 1 -1 1 0 0 -1 0.33Vbc 120 0.5773 -30 1.155 90 77 b c c c b c 1 0 0 -1 1 -1 0.33Vbc 0 0.5773 90 1.155 90 78 b c c c c c 1 0 0 0 0 -1 0.33Vbc 60 0.5773 30 1.155 90
79 b b b b b a 0 0 0 0 -1 1 0.33Vab 0 0.5773 -90 1.155 -30 80 b b b b a b 0 0 0 -1 1 0 0.33 Vab -60 0.5773 150 1.155 -30 81 b b b a b b 0 0 -1 1 0 0 0.33 Vab -120 0.5773 30 1.155 -30 82 b b a b b b 0 -1 1 0 0 0 0.33 Vab -180 0.5773 -90 1.155 -30 83 b a b b b b -1 1 0 0 0 0 0.33Vab 120 0.5773 150 1.155 -30 84 b b b a b a 0 0 -1 1 -1 1 0.33Vab -60 0.5773 -30 1.155 -30 85 b a b b b a -1 1 0 0 -1 1 0.33Vab 60 0.5773 -150 1.155 -30 86 b b a b a b 0 -1 1 -1 1 0 0.33Vab -120 0.5773 -150 1.155 -30 87 b a b a b b -1 1 -1 1 0 0 0.33Vab -180 0.5773 90 1.155 -30 88 b b a b a a 0 -1 1 -1 0 1 0.33Vab -60 0.9999 -120 1.155 150 89 b a b b a a -1 1 0 -1 0 1 0.33Vab 0 0.9999 -180 1.155 150 90 b a b a a b -1 1 -1 0 1 0 0.33Vab -120 0.9999 120 1.155 150 91 b b a a b a 0 -1 0 1 -1 1 0.33 Vab -120 0.9999 -60 1.155 150 92 b a a b b a -1 0 1 0 -1 1 0.33Vab 120 0.9999 -120 1.155 150 93 b a a b a b -1 0 1 -1 1 0 0.33Vab -180 0.9999 -180 1.155 150 94 b a b a a a -1 1 -1 0 0 1 0.33Vab -60 0.5773 150 1.155 150 95 b a a a b a -1 0 0 1 -1 1 0.33Vab -180 0.5773 -90 1.155 150 96 b a a a a a -1 0 0 0 0 1 0.33Vab -120 0.5773 -150 1.155 150
97 c c c c c a 0 0 0 0 1 -1 0.33Vca 180 0.5773 90 1.155 30 98 c c c c a c 0 0 0 1 -1 0 0.33Vca 120 0.5773 -30 1.155 30 99 c c c a c c 0 0 1 -1 0 0 0.33Vca 60 0.5773 -150 1.155 30
100 c c a c c c 0 1 -1 0 0 0 0.33Vca 0 0.5773 90 1.155 30 101 c a c c c c 1 -1 0 0 0 0 0.33Vca -60 0.5773 -30 1.155 30 102 c c c a c a 0 0 1 -1 1 -1 0.33Vca 120 0.5773 150 1.155 30 103 c a c c c a 1 -1 0 0 1 -1 0.33Vca -120 0.5773 30 1.155 30 104 c c a c a c 0 1 -1 1 -1 0 0.33Vca 60 0.5773 30 1.155 30 105 c a c a c c 1 -1 1 -1 0 0 0.33Vca 0 0.5773 -90 1.155 30 106
c c a c a a 0 1 -1 1 0 -1
0.33Vca 120 0.9999 60 1.155 -
150 107
c a c c a a 1 -1 0 1 0 -1
0.33Vca 180 0.9999 0 1.155 -
150 108
c a c a a c 1 -1 1 0 -1 0
0.33Vca 60 0.9999 -60 1.155 -
150 109
c c a a c a 0 1 0 -1 1 -1
0.33Vca 60 0.9999 120 1.155 -
150 110
c a a c c a 1 0 -1 0 1 -1
0.33Vca -60 0.9999 60 1.155 -
150 111
c a a c a c 1 0 -1 1 -1 0
0.33Vca 0 0.9999 0 1.155 -
150 112
c a c a a a 1 -1 1 0 0 -1
0.33Vca 120 0.5773 -30 1.155 -
150 113
c a a a c a 1 0 0 -1 1 -1
0.33Vca 0 0.5773 90 1.155 -
150 114
c a a a a a 1 0 0 0 0 -1
0.33Vca 60 0.5773 30 1.155 -
150
115 c c c c c b 0 0 0 0 -1 1 0.33Vbc 0 0.5773 -90 1.155 90 116 c c c c b c 0 0 0 -1 1 0 0.33Vbc -60 0.5773 150 1.155 90 117 c c c b c c 0 0 -1 1 0 0 0.33Vbc -120 0.5773 30 1.155 90 118 c c b c c c 0 -1 1 0 0 0 0.33Vbc -180 0.5773 -90 1.155 90 119 c b c c c c -1 1 0 0 0 0 0.33Vbc 120 0.5773 150 1.155 90 120 c c c b c b 0 0 -1 1 -1 1 0.33Vbc -60 0.5773 -30 1.155 90 121 c b c c c b -1 1 0 0 -1 1 0.33Vbc 60 0.5773 -150 1.155 90 122 c c b c b c 0 -1 1 -1 1 0 0.33Vbc -120 0.5773 -150 1.155 90 123 c b c b c c -1 1 -1 1 0 0 0.33Vbc -180 0.5773 90 1.155 90 124 c c b c b b 0 -1 1 -1 0 1 0.33Vbc -60 0.9999 -120 1.155 -90 125 c b c c b b -1 1 0 -1 0 1 0.33Vbc 0 0.9999 -180 1.155 -90 126 c b c b b c -1 1 -1 0 1 0 0.33Vbc -120 0.9999 120 1.155 -90 127 c c b b c b 0 -1 0 1 -1 1 0.33Vbc -120 0.9999 -60 1.155 -90 128 c b b c c b -1 0 1 0 -1 1 0.33Vbc 120 0.9999 -120 1.155 -90 129 c b b c b c -1 0 1 -1 1 0 0.33Vbc -180 0.9999 -180 1.155 -90 130 c b c b b b -1 1 -1 0 0 1 0.33Vbc -60 0.5773 150 1.155 -90 131 c b b b c b -1 0 0 1 -1 1 0.33Vbc -180 0.5773 -90 1.155 -90 132 c b b b b b -1 0 0 0 0 1 0.33Vbc -120 0.5773 -150 1.155 -90
133
133 c c c c b b 0 0 0 -1 0 1 0.5773 Vbc -30 0.5773 -150 2 90 134 c c c b b c 0 0 -1 0 1 0 0.5773 Vbc -90 0.5773 90 2 90 135 c c b b c c 0 -1 0 1 0 0 0.5773 Vbc -150 0.5773 -30 2 90 136 c b b c c c -1 0 1 0 0 0 0.5773 Vbc 150 0.5773 -150 2 90 137 c c b b b b 0 -1 0 0 0 1 0.5773 Vbc -90 0.5773 -90 2 -90 138 c b b b b c -1 0 0 0 1 0 0.5773 Vbc -150 0.5773 150 2 -90 139 b b a a a a 0 -1 0 0 0 1 0.5773 Vab -90 0.5773 -90 2 150 140 b a a a a b -1 0 0 0 1 0 0.5773 Vab -150 0.5773 150 2 150 141 c c c c a a 0 0 0 1 0 -1 0.5773 Vca 150 0.5773 30 2 30 142 c c c a a c 0 0 1 0 -1 0 0.5773 Vca 90 0.5773 -90 2 30 143 c c a a c c 0 1 0 -1 0 0 0.5773 Vca 30 0.5773 150 2 30 144 c a a c c c 1 0 -1 0 0 0 0.5773 Vca -30 0.5773 30 2 30 145
c c a a a a 0 1 0 0 0 -1
0.5773 Vca 90 0.5773 90 2 -
150 146
c a a a a c 1 0 0 0 -1 0
0.5773 Vca 30 0.5773 -30 2 -
150 147 b b b b a a 0 0 0 -1 0 1 0.5773 Vab -30 0.5773 -150 2 -30 148 b b b a a b 0 0 -1 0 1 0 0.5773 Vab -90 0.5773 90 2 -30 149 b b a a b b 0 -1 0 1 0 0 0.5773 Vab -150 0.5773 -30 2 -30 150 b a a b b b -1 0 1 0 0 0 0.5773 Vab 150 0.5773 -150 2 -30 151 b b c c c c 0 1 0 0 0 -1 0.5773 Vbc 90 0.5773 90 2 90 152 b c c c c b 1 0 0 0 -1 0 0.5773 Vbc 30 0.5773 -30 2 90 153 a a c c c c 0 -1 0 0 0 1 0.5773 Vca -90 0.5773 -90 2 30 154 a c c c c a -1 0 0 0 1 0 0.5773 Vca -150 0.5773 150 2 30 155 b b b b c c 0 0 0 1 0 -1 0.5773 Vbc 150 0.5773 30 2 -90 156 b b b c c b 0 0 1 0 -1 0 0.5773 Vbc 90 0.5773 -90 2 -90 157 b b c c b b 0 1 0 -1 0 0 0.5773 Vbc 30 0.5773 150 2 -90 158 b c c b b b 1 0 -1 0 0 0 0.5773 Vbc -30 0.5773 30 2 -90 159
a a a a c c 0 0 0 -1 0 1
0.5773 Vca -30 0.5773 -150 2 -
150 160 a a a a b b 0 0 0 1 0 -1 0.5773 Vab 150 0.5773 30 2 150 161 a a a b b a 0 0 1 0 -1 0 0.5773 Vab 90 0.5773 -90 2 150 162 a a b b a a 0 1 0 -1 0 0 0.5773 Vab 30 0.5773 150 2 150 163 a b b a a a 1 0 -1 0 0 0 0.5773 Vab -30 0.5773 30 2 150 164 a a b b b b 0 1 0 0 0 -1 0.5773 Vab 90 0.5773 90 2 -30 165 a b b b b a 1 0 0 0 -1 0 0.5773 Vab 30 0.5773 -30 2 -30 166
a a a c c a 0 0 -1 0 1 0
0.5773 Vca -90 0.5773 90 2 -
150 167
a a c c a a 0 -1 0 1 0 0
0.5773 Vca -150 0.5773 -30 2 -
150 168
a c c a a a -1 0 1 0 0 0
0.5773 Vca 150 0.5773 -150 2 -
150 169 a a a b b b 0 0 1 0 0 -1 0.6666Vab 120 0 160.0 2.31 -30 170
a a b b b a 0 1 0 0 -1 0
0.6666Vab 60 0 -
141.3 2.31 -30
171 a b b b a a 1 0 0 -1 0 0 0.6666Vab 0 0 90 2.31 -30 172 a a c c c a 0 -1 0 0 1 0 0.6666 Vca -120 0 38.66 2.31 30 173
a a a c c c 0 0 -1 0 0 1
0.6666 Vca -60 0 -
19.98 2.31 30
174 a c c c a a -1 0 0 1 0 0 0.6666 Vca -180 0 -90 2.31 30 175 b b b c c c 0 0 1 0 0 -1 0.6666 Vbc 120 0 160.0 2.31 90 176
b b c c c b 0 1 0 0 -1 0
0.6666 Vbc 60 0 -
141.3 2.31 90
177 b c c c b b 1 0 0 -1 0 0 0.6666 Vbc 0 0 90 2.31 90 178
b b b a a a 0 0 -1 0 0 1
0.6666Vab -60 0 -
19.98 2.31 150
179 b b a a a b 0 -1 0 0 1 0 0.6666Vab -120 0 38.66 2.31 150 180 b a a a b b -1 0 0 1 0 0 0.6666Vab -180 0 -90 2.31 150 181
c c c a a a 0 0 1 0 0 -1
0.6666 Vca 120 0 160.0 2.31 -
150 182
c c a a a c 0 1 0 0 -1 0
0.6666 Vca 60 0 -
141.3 2.31 -
150 183
c a a a c c 1 0 0 -1 0 0
0.6666 Vca 0 0 90 2.31 -
150 184
c c c b b b 0 0 -1 0 0 1
0.6666 Vbc -60 0 -
19.98 2.31 -90
185 c c b b b c 0 -1 0 0 1 0 0.6666 Vbc -120 0 38.66 2.31 -90 186 c b b b c c -1 0 0 1 0 0 0.6666 Vbc -180 0 -90 2.31 -90
134
All the 189 space voltage vectors can be summarized in Table 5.5.
Table 5.5. The 189 vectors of table.5.4 are distributed as:
Space vectors No of vectors Magnitude of Vdq
Magnitude of Vxy
Magnitude of V0+0-
Input current
Corresponding to output voltage
1-24 18 0 1.1546 0 0
6 0 0 1.9998 0
25-132 72 0.3333 0.5773 0.6666 1.155
36 0.3333 0.9999 1.13332 1.155
132-168 36 0.5773 0.5773 0.000 2
169-186 18 0.6666 0.000 0.6666 2.31
186-189 3 0 0 0 0
The space voltage vectors are shown in Fig. 5.10. There are several redundant voltage
vectors. Hence at each location several vectors are located.
Fig. 5.10. Three-phase to six-phase DMC output voltage (Vdq).
135
Fig. 5.10a. Three-phase to six-phase DMC output voltage (Vxy).
Fig. 5.10b. Three-phase to six-phase DMC output voltage (Vo+,o-).
Fig. 5.10c. Three-phase to six-phase DMC input current.
136
5.4 Space Vector Model of Three-phase to Seven-phase Direct Matrix Converter
The topology of the Matrix Converter elaborated in this thesis is fed using a standard three-
phase supply and outputs seven-phase. Each leg has three bi-directional power semi-
conductor switches and as such twenty one switches are used. A three-to-seven phase Matrix
Converter can be represented in two ways: direct and indirect. The direct Matrix Converter
considers the whole circuit as one entity and on the other hand indirect Matrix Converter is
assumed as a combination of a rectifier and an inverter fictitious DC link. A general topology
of a three-phase to seven-phase DMC is shown in Fig 5.11.
Fig. 5.11. Three-phase to seven-phase direct Matrix Converter.
For a balanced seven phase sinusoidal system the instantaneous voltages maybe expressed as;
7/12
7/10cos
7/8cos
7/6cos
7/4cos
7/2cos
cos
0
tco
t
t
t
t
t
t
V
V
V
V
V
V
V
V
o
o
o
o
o
o
o
G
F
E
D
C
B
A
(5.10)
In terms of space vector representation it can be expressed as:
)(7
2 654320 gFEDCBA VaVaVaVaVaaVVV (5.11)
7/2jea (Seventh root of unity) and 2/7 is a scaling factor equal to the ratio between the
magnitude of the output line-to-line voltage and that of output voltage vector. The angular
velocity of the vector is o and its magnitude V0.
Similarly, the space vector representation of the three-phase input voltage is given by;
)(3
2 2cbai VaaVVV (5.12)
137
3/2jea (cube root of unity) and 2/3 is a scaling factor equal to the ratio between the
magnitude of the input line-to-line voltage and that of input voltage vector.
If a balanced seven-phase load is connected to the output terminals of the direct Matrix
Converter, the space vector forms of the seven phase output currents and input currents are
given by:
)(7
2 654320 GFEDCBA IaIaIaIaIaaIII
)( ootj
oeI (5.13)
))()()((
3
2 2 tIataItII cbai (5.14)
)( iotj
ieI
where, O is the phase shift angle of the output current to the output voltage and i is that of
the input current to the input voltage. There are twenty one power semiconductor switches
and each switch is free to connect any of the three phases of input (Fig 5.11). In this way
there are 3*7=21 possible equivalent switch. Every switch can have two states ‘on’ (1)or ‘off’
(0). In this way there are 221 switching combinations. Considering the constraints “that any
two input cannot be shorted and output phase cannot be opened throughout the operation”
switching combination reduces to 37 = 2187 only. These 2187 switching combinations are
analyzed in eight groups that shows how many output phases are connected from input phase
(a, b, c). It is represented as x,y,z where x, y and z are no of output phases connected to
input phase a, b and c, respectively. These groups are elaborated below:
Group-1, [7,0,0]:- This group encompasses those switching vectors that are produced when
all seven output phases are connected with the input phase ‘a’. Hence, as such there exist two
more situations, where all the output phases are connected to phase ‘b’ and all output phases
are connected to phase ‘c’ and they are denoted as 7,0,0, 0,7,0,0,0,7. Here 7,0,0,
means all seven output phases are connected to phase ‘a’ of input. Hence, three possible
switching states are available.
Group-2, [6,0,1]:- In this group 6 output phases are connected to the same input phase and
remaining phase is connected to any of the remaining input phases. Thus there are 7C6 (6 out
of 7 output phases are connected to same input) and 1C1 (remaining one output from any one
input).
(7!/6!*1!) = 7 combinations , In this way the possible switching combination sub-groups are;
138
6,0,1,6,1,0,0,6,1,1,6,0,0,1,6,1,0,6. One sub group can have (7!/6!*1!) = 7
combinations. As such there are 6 sub-groups and thus the total possible combination will be
7*6 = 42 switching combinations.
Group-3, [5,0,2]:-This group encompasses those switching combination where 5 output
phases are connected to one input phase and the rest two outputs are connected to any other
remaining input phases. The nomenclature 5,0,2 indicates that five output phases are
connected to input phase ‘a’, no output phase are connected to input phase ‘b’ and two output
phases are connected to input phase ‘c’. The other possible combinations are:
,5,2,0,0,5,2,2,5,0,0,2,5,2,0,5. Thus this group yield:
7C5 *2C2 =(7!/5!*2!) * (2!/2!*0!) =21*6 = 126 possible switching states and hence 126 space
vectors.
Group-4,[5,1,1]:- This group present those switching states where five output phases are
connected to one input phase and the rest two output phases are connected to two inputs. As
such three possible situation can occur 5,1,1,1,5,1,1,1,5. The total number of possible
switching states and hence space vectors are: 7C5 *
2C1*1C1 =(7!/5!*2!) * (2!/1!*1!) =21*2*3 = 126.
Group-5, [4,0,3]:- This group is comprised of those states that are produced when four
output phases are connected to one input phase and the remaining three output phases are
connected to any of the two available input phases. Hence the possible situations are 4,0,3,
4,3,0, 0,4,3, 3,4,0, 0,3,4, 3,0,4. The total number of space vectors yield are: 7C4 *
3C3 = (7!/4!*3!) * (3!/3!*0!) =35*6 = 210.
Group-6, [4,2,1]:-This group yield largest number of switching combinations. The possible
cases are such that four output phases are connected with one input phase, two output phases
are connected to another input phase and one output is connected with one input. Hence the
complete sets are 4,2,1, 4,1,2,1,4,2,2,4,1,1,2,4,2,1,4. The possible number of
switching combination and hence space vectors are obtained by the following combinational
rule: 7C4 *
3C2*1C1 =(7!/4!*3!) * (3!/3!*0!) =35*3*6 = 630.
139
Group-7, [3,2,2]:- This group also encompasses largest number of possible space vectors.
The set comprised of situation where three output phases are connected to one input, the
remaining two outputs each are connected to the remaining one input each. Thus the possible
sets are 3,2,2,2,3,2,2,2,3. The total number of possible space vectors are: 7C3 *
4C2 *2C2 =(7!/4!*3!) * (4!/2!*2!) =35*6*3 = 630.
Group-8,[3,1,3]:- This group is second largest. The possible sets are when three output
phases each are connected to one input phase each and the remaining one output phase is
connected to one remaining input phase. The possible group subsets are 3, 13, 3, 3, 1,
1,3,3. The possible switching states and hence space vector are:
7C3 *4C3 *
1C1 =(7!/4!*3!) * (4!/3!*1!) =35*4*3 = 420.
There are in total = 3 + 42 + 126 + 126 + 210 + 630 + 420 + 630 = 2187 switching
combinations in which group [1], [2], [3] and [5] are having varying amplitude and fixed
frequency. Such switching combinations are 3 + 42 + 126 + 210 = 381. Remaining are having
varying amplitude and frequency both so they are not helpful to realize space vector
modulation. These 381 switching states that are realizable using space vector PWM is given
in Table.5.5. The total 381 vectors are then classified or sub-grouped of 42 vectors each
according to their position and magnitude. This table shows the output line voltage of the
Matrix Converter having lowest phase difference between phase voltages i.e. adjacent line
voltages, VAB ,VBC, VCD, VDE, VEF, VFG, VGA. The other sets of line voltages are possible in a
seven-phase system that are called non-adjacent-1 and non-adjacent-2 but they are not
considered in this discussion. The voltages can be transformed into three sets of orthogonal
planes in a seven dimensional space namely Vdq, Vx1y1 and Vx2y2 and are obtained by using
the following transformation equations (5.15) given in reference [5.5],
)(7
2 65432GAFGEFDECDBCABdq VaVaVaVaVaaVVV
)(7
2 1210864211 GAFGEFDECDBCAByx VaVaVaVaVaVaVV
)(7
2 18151296322 GAFGEFDECDBCAByx VaVaVaVaVaVaVV (5.15)
Where 7/2jea , the space vectors of x1-y1 planes represent the third harmonic of d-q and
x2-y2 represent fifth harmonics. The space vectors obtained are illustrated in Figs. 5.12-5.14.
140
The zero sequence vectors are not considered as the load assumed is with isolated neutral
point.
The space vectors are generated in pairs such that they have equal magnitude but opposite
polarity (180 phase shift), hence they are denoted by + and – signs. Further they appear in
terms of Vab, Vbc and Vca (input line voltages).
Table.5.5. Output voltage space vectors in d-q plane.
Groups number of vectors
vectors Magnitude
[G1] 3 1 to 3 0 14
)12(
,, *
kj
cabcab eV
[G2] 21 +1 to +21 0.11 14
)12(
,, *
kj
cabcab eV
[G3] 21 +1 to +21 0.14 14
)12(
,, *
kj
cabcab eV
[G4] 21 +1 to +21 0.20 14
)12(
,, *
kj
cabcab eV
[G5] 21 +1 to +21 0.25 14
)12(
,, *
kj
cabcab eV
[G6] 21
+1 to +21 0.31 14
)12(
,, *
kj
cabcab eV
[G8] 21
(Medium)
+1 to +21 0.45 14
)12(
,, *
kj
cabcab eV
[G9] 21
(Large)
+1 to +21 0.56 14
)12(
,, *
kj
cabcab eV
These 383 active vectors are named on the basis of magnitude and position from ±1 to ±21.
Hence there are But in the Table (5.5) only +1 to +21 are shown with their magnitude and
direction, remaining -1 to -21 is not shown as those are completely same but 180o phase
opposition of +1to +21. Group seven is not shown in the Table. It has 84 vectors these are
141
unequally spaced but of same magnitude and these vectors are not considered for voltage and
current analysis. The group vectors G[8] and G[9] are further utilized in implementing space
vector PWM since they have higher magnitude.
The space vector diagram has 3 space vectors at each locations and hence there are only 127
(383/3) location as shown in the Figs. 5.12 - 5.15. It is to be noted that the total space vectors
are encompassed within a circle with sub division of 14 major sectors (considering only large
length vectors). However, within the larger sectors there are two distinct positions of space
vectors. On the large length there are seven vectors lying on the same line but with different
switching combinations.
Fig. 5.12. Adjacent line space vectors in d-q plane.
142
Fig. 5.13. Adjacent Line voltage space vectors in x1-y1 plane.
0.2
0.4
0.6
0.8
1
30
210
60
240
90
270
120
300
150
330
180 0
143
Fig. 5.14. Adjacent Line voltage space vectors in x2-y2 plane.
0.5
1
1.5
30
210
60
240
90
270
120
300
150
330
180 0
144
Fig. 5.15. Three phase source current for 3-7 phase MC.
5.5 Summary
This chapter developed complete space vector model of three configurations of Matrix Converter:
Three-phase input Five-phase output
Three-phase input Six-phase output
Three-phase input Seven-phase output.
In case of five-phase output the number switching combinations are215= 32768. Out of these
many switching combinations not all of them are useful. When choosing practically possible
switching combination, the safe operation of Matrix Converter is to be considered. The input
side should not be short circuited an the output side should not be open circuited. When
considering these constraints the possible number of switching combination reduces to 53 =
243. It is seen that the number of switching combinations is same as that of a three-level five-
phase inverters. Therefore, one can say that the operation of a Matrix Converter is similar to
the operation of a three-level back-to-back converter. These switching combinations produce
space vectors that are grouped appropriately.
In case of three-phase input and six-phase output the total possible numbers of switching
combinations are 218=262144. Once again the switching combinations are reduced by
considering the safety of operation. The actual number of permissible switching combinations
are36 = 729. They produced the same number of space vectors that can be grouped according
to their magnitude.
145
In case of three-phase input and seven-phase output the total possible numbers of switching
combinations are 221 =262144. Once again the switching combinations are reduced by
considering the safety of operation. The actual number of permissible switching combinations
are36 = 729. They produced the same number of space vectors that can be grouped according
to their magnitude.
In case of three-phase input and seven-phase output there are 221 = 2097152 possible
switching combinations. Considering the constraints “that any two input cannot be shorted
and output phase cannot be opened throughout the operation” switching combination reduces
to 37 = 2187 only. These produce 2187 switching combinations. The sorting and grouping of
vectors are reported.
References:
[5.1]. SK. M. Ahmed, A. Iqbal, H. Abu-Rub, M. Saleh, A. Kalam. “Vector Control of a Five-phase induction motor supplied by a Non-square 3X5 phase Matrix Converter,” Australian Journal of Electrical Engineering, vol. 10, no. 1, March 2013, pp. 55-63.
[5.2]. SK. M. Ahmed, H. Abu-Rub, A. Iqbal, M.A. S. Payami, “A three-to-five-phase Matrix Converter based five-phase induction motor drive system,” Int. Journal of Recent Trend in Engg. & Tech., vol. 8, no. 2, Jan 2013, pp. 82-86. DOI: 01.IJRTET.8.2.94_503.
[5.3]. A. Iqbal, SK. M. Ahmed, H. Abu-Rub, “Space vector PWM technique for a three to five-phase Matrix Converter,” IEEE Tran. On Ind. Appl. Vol. 48, No. 2, March/April 2012, pp. 697-707.
[5.4]. H. Abu-Rub, A. Iqbal, J. Guzisnki, “High Performance control of AC Drives With MATLAB/SIMULINKModels”, Wiley UK, 2012.
[5.5]. E. Levi, “Multi-phase Machines for Variable speed applications” IEEE Trans. Ind. Elect., vol. 55, no. 5, pp. 1893-1909, May 2008.
146
Chapter 6 Carrier Based PWM Schemes for Feeding
Multi-Motor Drive System
6.1 Introduction
This chapter aims to investigate control algorithm of a Matrix Converter. Multi-phase Matrix
Converters are discussed in detail with the help of MATLAB simulation. The presentation in
this thesis started from three to three-phase Matrix Converter through the control of three to
five, three to six and three to seven phase configurations.
An AC to AC power converter is widely used in high-performance drive systems due to its
bi-directional power flow and unity input power factor operation capabilities. It allows a wide
range of output voltage. A conventional AC-AC converter consists of pulse width modulation
(PWM) buck converter and a PWM inverter with DC voltage link. As an energy storage
component, it requires a large capacitor in the DC link and a bulky inductor at the input
terminal. The DC-link capacitor can be a critical component, especially in high-power or
high-voltage applications, since it is large and expensive, and it has a limited lifetime. The
source-side inductors are also a burden to the system. An AC/AC converter with bi-
directional switches and no reactive components are implemented to overcome all listed
disadvantage of conventional AC/AC converters.
Controlled bi-directional power flow using direct AC-AC conversion using semi-conductor
switches arranged in the form of matrix array, popularly known as Matrix Converter. Matrix
Converter fed motor drive is superior to voltage source inverter because of its bi-directional
power switching characteristic and compact design without bulky capacitors. Even though,
Matrix Converter is used less in industries due to its complex switching design and
modulation techniques. Modulation methods of Matrix Converters are complex and are
generally classified in two different groups, called direct and indirect. The direct PWM
method developed by Alesina and Venturini in reference [1.18] limits the output to half the
input voltage. This limit was subsequently raised to 0.866 by taking advantage of third
harmonic injection and it was realized that this is maximum output that can be obtained from
a three-to-three phase Matrix Converter in the linear modulation region. Indirect method
assumes a Matrix Converter as a cascaded virtual three-phase rectifier and a virtual voltage
147
source inverter with imaginary DC link. With this representation, space vector PWM method
of VSI is extended to a Matrix Converter. Although the space vector PWM method is suited
to three-phase system but the complexity of implementation increases with the increase in the
number of switches/phases. Motivated from the simple implementation, carrier-based PWM
scheme has been introduced recently for three to three phase Matrix Converter. The same is
extended in this thesis for multi-phase Matrix Converter. The investigated topologies are for
single-motor drive and two-motor drive.
6.2 Generalized Carrier-based PWM Techniques for three-to N-phase
In this section generalised 3xN-phase Matrix Converter is discussed and developed scheme is
modular in nature and is thus applicable to the generalised circuit topology of Fig 6.1 which
shows a Matrix Converter that produces ‘N’ phase output with three sinusoidal input and
‘Nx3’ array of switching circuit.
Fig. 6.1. Power circuit topology of 3xN-phase Matrix Converter.
The load to the Matrix Converter is assumed as star-connected N-phase AC machine.
Switching function is defined as:
Sjk= 1 for closed switch, 0 for open switch, j = a,b,c (input), k= A,B,C,D,E (output).
The switching constraint is:
148
Sak+ Sbk+ Sck = 1 (6.1)
The modulation technique is developed by assuming input side as three-phase controlled
rectifier and the output is N -phase voltage source inverter with a fictitious DC link.
Let ‘ )(tKJ ’ as duty ratio
and We can define as offon
onKJ t
)(
(6.2)
where 0≤ )(tKJ ≤ 1
The low frequency components of output phase voltage
(6.3)
where, iV is the instantaneous input voltage vector. D(t) is the low frequency transfer Matrix
Converter and can be defined as;
)()()(
)()()(
)()()(
)(
)(
ttt
ttt
ttt
ttt
tD
cNbNaN
cCbCaC
cBbBaB
cAbAaA
(6.4)
In a 3xN Matrix Converter the input three levels are cba VVV ,, . The output levels are
NCBA VVVV ...,, Output waveforms of the matrix converter are generated by a small portion of
the input waveform with respect to the control signal. Equation (6.5) shows a switching
matrix formed by duty ratio of the 3xN Matrix Converter.
c
b
a
cNbNaN
cCbCaC
cBbBaB
cAbAaA
N
C
B
A
V
V
V
V
V
V
V
(6.5)
And input current:
)().()(0 tVtDtV i
149
N
C
B
A
cNbNaN
cCbCaC
cBbBaB
cAbAaA
c
b
a
I
I
I
I
I
I
I
(6.6)
In this chapter a balanced three-phase system is considered in the input side. The input
voltages are given as:
sin( ),
sin( 2 / 3),
sin( 4 / 3)
a
b
c
v V t
v V t
v V t
(6.7)
The output voltage duty ratios should be calculated in such a way that output voltages
remains independent of input frequency. In a different way, the output voltages can be
considered in synchronous reference frame and the three-phase input voltages can be
considered to be in stationary reference frame, so that the input frequency term will be absent
in output voltages. Considering the above points in mind duty ratios of thM output phase are
chosen as given in references [6.1-6.2]:
)3/4cos(
),3/2cos(
),cos(
tkd
tkd
tkd
McM
MbM
MaM
(6.8)
Therefore the thM phase output voltage can be obtained by using the above duty ratios as
)]3/4cos()3/4cos(
)3/2cos()3/2cos()cos()[cos(
tt
ttttVkv MA
(6.9)
Equation (6.9) can be also written as:
)cos(2
3 Vkv MA (6.10)
In equation (6.10), )cos( term indicates that the output voltage is affected by . Thus, the
output voltage Mv is independent of the input frequency and only depends on the amplitude
V of the input voltage and Mk is a reference output voltage time-varying modulating signal
for the thM output phase with the desired output frequency o .The N-phase reference output
voltages can be represented as:
150
)/)1(2cos(
),/2cos(
),cos(
nntkk
ntkk
tkk
on
oB
oA
(6.11)
Therefore, from equation (6.10), the output voltage in phase-M is:
)cos()cos(2
3tVkv oMA
(6.12)
6.2a Application of offset duty ratios
While considering any of the phase output the duty ratio will become negative. To overcome
this ‘n’ offset duty ratios are added to the all phases equally which lies between zero and one
in all instances. These offset voltages will appear as a common mode voltage in output and
nullify in output phases.
Thus offset duty ratio will be added to final duty ratio of any phase output as shown in
equation (6.13).
(6.13)
The sum of input duty ratio is zero at any instant. Thus additional offset duty ratio can be
calculates as
(6.14)
To make unity output, this duty ratio is added to any of the duty ratio such as
)()(),( tandDtDtD cba . . When added to three duty ratio the offset duty ratio will become:
(6.15)
The maximum value of K is 0.5. So total new duty ratio after adding offset will become:
(6.16)
( )aN aN ad D t
))()()(1( tDtDtD cba
3/))()()(1( tDtDtD cba
)3/4()()3/4(5.0
)3/2()()3/2(5.0
)()()(5.0
tCostKtCos
tCostKtCos
tCostKtCos
AcAcA
AbAbA
AaAaA
151
Here, in any of the cycle the output phase should not be open circuited. Thus the system
maintains output voltage and input current is unaffected. To utilise the maximum input
voltage capability an additional components is also used and it enhances the value of Ka, Kb,
Kc from 0.5 to 0.57.
i.e. ]2/)()( ,,,, CBACBA KKKMinKKKMax (6.17)
Thus the final duty ratio of output phase ‘a’ is:
tCosKKKKMinKKKMaxKtDtDtDtD
tCosKKKKMinKKKMaxKtDtDtDtD
tCosKKKKMinKKKMaxKtDtDtDtD
ACBACBAAcbaccA
ACBACBAAcbabbA
ACBACBAAcbaaaA
(]2/)()(3/))()()(1()(
)(]2/)()(3/))()()(1()(
)(]2/)()(3/))()()(1()(
,,,,,
,,,,,
,,,,,
(6.18)
The input current is represented as the sum of time weighting current and duty ratio. The
input current of phase ‘a’ can be represented as:
(6.19)
In the balanced three phase supply, the offset duty ratio present in final duty ratio will not
affect the input current. So
(6.20)
We can simplify the above equation (6.20) can be simplified as k equal to amplitude of the
modulation and 0I as the amplitude of the output current and 0 as the output power factor
angle.
i.e. CCBBAA
a
iKiKiKKI
CoskII
0
002
3
(6.21)
The switching signals are generated after comparing the input with a triangular waveform.
)3/4cos()(
),3/2cos()(
),cos()(
tktDd
tktDd
tktDd
MccM
MbbM
MaaM
(6.22)
In any switching cycle the output phase has to be connected to any of the input phases. The
summation of the duty ratios must be equal to unity. But the summation )()()( tDtDtD cba
is less than or equal to unity. Hence, another offset duty-ratio 3/)()()(1 tDtDtD cba is
CaCBaBAaAa iiiI
)()( tCosiKiKiKI CCBBAAa
152
added to ( ), ( ) ( )a b cD t D t and D t . The addition of this offset duty-ratio in all switches will
maintain the output voltages and input currents unaffected. Similarly, the duty-ratios are
calculated for the other output phases.
If nBA kkk ........., are chosen to be n-phase sinusoidal references as given in Equation. (6.17),
the input voltage capability is not fully utilized for output voltage generation. To overcome
this, an additional common mode term equal to
]2/),........,,min(),........,,max( nBAnBA kkkkkk (6.23)
is added as in the carrier-based space-vector PWM principle as implemented in two-level
inverters. Thus the amplitude of nBA kkk ........., can be enhanced from 0.5 with respect to the
number of output phases.
6.2b Without Common Mode Voltage Addition
The duty ratio for Mth output phase can be written as:
)3/4cos(3/))()()(1()(
)3/2cos(3/))()()(1()(
)cos(3/))()()(1()(
tktDtDtDtDd
tktDtDtDtDd
tktDtDtDtDd
McbaacM
McbaabM
McbaaaM
(6.24)
6.2.c With Common Mode Voltage Addition
The duty ratio for Mth output phase can be written as:
(6.25)
where
( ) (1 ( ) ( ) ( )) / 3
[ max( ,......., ) min( ,...... ) / 2] cos( )
( ) (1 ( ) ( ) ( )) / 3
[ max( ,......., ) min( ,...... ) / 2] cos( 2 / 3 )
( ) (1
aM a a b c
M A M A M
bM a a b c
M A M A M
cM a
d D t D t D t D t
k k k k k t
d D t D t D t D t
k k k k k t
d D t D
( ) ( ) ( )) / 3
[ max( ,......., ) min( ,...... ) / 2] cos( 4 / 3 )a b c
M A M A M
t D t D t
k k k k k t
( ) 0.5cos( )
( ) 0.5cos( 2 / 3 )
( ) 0.5cos( 4 / 3 )
a
b
c
D t t
D t t
D t t
153
(6.26)
6.2d Output Side over Modulation
Any increase from the maximum amplitude of the reference signal, reduces the fundamental
gain of the modulator. Six-step square wave operating mode gives the maximum fundamental
output voltage. The maximum possible magnitude of the factious DC voltage depends on the
peak input voltage amplitude and input power factor:
(6.27)
Maximum fundamental output voltage can be calculated as:
(6.28)
Equation (6.27) shows that over modulation in output side increases the input voltage transfer
ratio from 0.866 to 0.95. This increase in input voltage transfer ratio increases the low
frequency harmonics in the output voltage and input current.
6.2e Modulator Gain
The output side over modulation can be divided into two operating mode when using space-
vector modulation. The fundamental gain of the modulator as a function of peak amplitude
modulation can be expressed as shown in reference [6.3]
(6.28)
(6.29)
(6.30)
Where, fundAK _ is the fundamental component of the output modulating signal in the over-
modulation region and is equal to kA when operating in the linear modulation range.
6.3.1 Carrier Based PWM Technique for a Three-to-Five Phase Matrix Converter
In this section, a carrier based PWM strategy is presented based on the comparison of the
modulating signals (five-phase target output voltages) with the high frequency triangular
)cos(2
3max_ inFictious vV
inFict
o Vv
v32 max
21_
3
11
2
1
3
1sin
3
AAA
fundA
KKK
K
fictdcAfunda VKV __ *
154
carrier wave. The output voltage is limited to 0.75 of the input voltage magnitude. Another
scheme is suggested in this result utilising the injection of common mode voltage in the
output five-phase target voltage. This results in enhanced output voltage equal to 0.7886 of
the input magnitude. Theoretically, this is the maximum output magnitude that can be
obtained in this Matrix Converter configuration in the linear modulation region. Analytical
approach is used to develop and analyse the proposed modulation techniques and are further
supported by simulation and experimental results.
6.3a Three-To-Five Phase Matrix Converter
The power circuit topology of a three-phase to five-phase Matrix Converter is illustrated in
Fig. (6-2). There are five legs with each leg having three bi-directional power switches
connected in series. Each power switch is bi-directional in nature with anti-parallel connected
IGBTs and diodes. The input is similar to a three-phase to three-phase Matrix Converter
having LC filters and the output is five-phases with 72˚ phase displacement between each
phases. Diode clamped circuit is also used in the hardware to protect the power devices while
commutating. The input side uses three diodes while the output side uses five diodes for
clamping circuit.
Fig. 6.2. Power Circuit topology of three-phase to five-phase Matrix Converter.
The load to the Matrix Converter is assumed as star-connected five-phase AC machine. The
modulation technique is developed by assuming input side as three-phase controlled rectifier
and the output is a five-phase voltage source inverter with a fictitious DC link.
In a 3x5 Matrix Converter the input three levels are cba VVV ,, . The output five levels are
155
, , , ,A B C D EV V V V V . Output waveforms of the Matrix Converter are generated by a small
portion of the input waveform with respect to the control signal. Equation (6.31) shows a
switching matrix formed by duty ratio of the 3x5 Matrix Converter as shown in reference
[6.4].
A aA bA cA
aB a B bB cB
a C bC cCC b
a D b D a DD c
a E bE cEE
V
VV
V V
V V
V
(6.31)
The input current
A aA bA cA
aB aB bB cB
aC bC cCC b
aD bD aDD c
aE bE cEE
I
II
I I
I I
I
(6.32)
The balanced three system is considered as input; so the input voltage is:
sin( ),
sin( 2 / 3),
sin( 4 / 3)
a
b
c
v V t
v V t
v V t
(6.33)
The output voltage duty ratios should be calculated such as output frequency should be independent of input frequency. So that input frequency term will be absent in output voltages. Hence, the duty ratios of output phase are chosen as:
cos( ),
cos( 2 / 3 ),
cos( 4 / 3 )
aA A
bA A
cA A
d k t
d k t
d k t
(6.34)
Therefore the output voltage of phase ‘A’ can be written as:
3cos( )
2A Av k V
(6.35)
The five-phase reference output voltages can be written as:
(6.36)
)5/8cos(
)5/6cos(
)5/4cos(
)5/2cos(
)cos(
tmk
tmk
tmk
tmk
tmk
oE
oD
oC
oB
oA
156
where is input frequency and o is the output frequency, m is the modulation index. For
unity power factor has to be chosen as zero.
The output voltage can be computed as:
)5
8cos()cos(2
3
)5
6cos()cos(2
3
)5
4cos()cos(2
3
)5
2cos()cos(2
3
)cos()cos(2
3
tVkV
tVkV
tVkV
tVkV
tVkV
oEE
oDD
oCC
oBB
oAA
(6.37)
6.3b Simulation Results
Simulation results are shown in Figs. 6.3 and 6.4 for the modulation without common mode
voltage addition in the output target voltage. The results with common mode voltage addition
will remain the same except with the enhanced output magnitude. For simulation purpose,
switching frequency is kept at 6 kHz and fundamental frequency is chose as 50 Hz. The input
voltages are kept at 100 V peak. The spectrum of source current shows only fundamental
component, that indicates sinusoidal source current. The source signal contains switching
ripple that is filtered out using small LC filter at the input side.
a. b.
Fig. 6.3. Input side waveforms of 3 to 5-phase Matrix Converter: (a) Input voltage and current (b) Spectrum Input current.
0 0.02 0.04 0.06 0.08 0.1 0.12-100
-50
0
50
100
Time(sec)
Inp
ut v
olta
ge
& c
urr
en
t (V
,A)
Current
Voltage
0 0.02 0.04 0.06 0.08 0.1 0.12-20
0
20
Inp
ut C
urre
nt(A
)
Time (sec)
0 50 200 400 600 800 1,0000
5
10
15
Frequency(Hz)Cur
rent
Spe
ctru
m(A
)
157
a.
b.
Fig. 6.4. Output side waveforms of 3 to 5-phase Matrix Converter: (a) Five-phase filtered output phase voltages (b) Spectrum output voltage.
6.3.2 Career Based PWM Technique for a Three-to-Five Phase Matrix Converter for Supplying Five-phase Two-motor Drives
This section focuses on the development of a topology of Matrix Converter which is a single
stage power converter to produce more than three phases supplying five-phase two-motor drive
system. Here, a carrier based PWM strategy is presented based on the comparison of the
modulating signals (five-phase two-frequency target output voltages) with the high frequency
triangular carrier wave for a three to five-phase Matrix Converter. The output voltage is
limited to 0.75 of the input voltage magnitude if common-mode voltage is not injected.
Another scheme is utilising the injection of common mode voltage in the output five-phase
target voltage. This results in enhanced output voltage which is equal to 0.7886 of the input
magnitude. Theoretically, this is the maximum output magnitude that can be obtained in this
Matrix Converter configuration in the linear modulation region. Analytical approach is used to
develop and analyse the proposed modulation techniques and are further supported by
simulation results. The major aim of the thesis is to produce two fundamental frequency
output from the Matrix Converter that can be used to control two series/parallel connected
five-phase machines independently.
0 0.02 0.04 0.06 0.08 0.1-80
-60
-40
-20
0
20
40
60
80
Time (sec)
Out
put F
ilter
ed P
hase
Vol
tage
s (V
)
0 0.02 0.04 0.06 0.08 0.1 0.12-200
0
200
Out
put V
olta
ge(V
)
Time (sec)
25 200 400 600 800 1000
50
100
0
Frequency (Hz)
Vol
tage
Spe
ctru
m (
V)
158
6.4a Five-Phase Two-Motor Drive System
For the sake of completeness of the discussion, a brief description is presented for five-phase
two-motor drive. As mentioned in Chapter 4, in five-phase system two set of orthogonal
voltage/current components are produced namely d-q and x-y. In single-motor drive system,
only d-q components are utilised and the x-y components are free to flow creating losses.
Thus concept of two-motor five-phase drive system is developed where both these
components are utilised, d-q by one machine and x-y by other machine. The extra set of
current components (x-y) available in a five-phase system is effectively utilised
independently, controlling an additional five-phase machine when the stator windings of two
five-phase machines are connected in series (Fig. 6.5)/parallel (Fig. 6.6) and are supplied
from a single five-phase VSI. In Fig. 6.5, reference currents generated by two independent
vector controllers, are summed up as per the transposition rules and are supplied to the series-
connected five-phase machines. As such the two five-phase machines are supplied from one
five-phase inverter, but are controlled independently. More detail on this configuration of the
drive system is available in references [6.5]. The voltage and current relationship of this drive
topology is shown in equation (6.38).
21
21
21
21
21
desE
bdsD
ecsC
cbsB
aasA
vvv
vvv
vvv
vvv
vvv
21
21
21
21
21
desE
bdsD
ecsC
cbsB
aasA
iii
iii
iii
iii
iii
(6.38)
Here ‘s’ stands for source quantity, capital letters denote source and small letters represents
motor side quantities.
Another analogue stator winding connection of five-phase two-motor drive is parallel-
connected drive (Fig.6.6) which is directly derived from the analogy of series and parallel
circuits. The source (three to five-phase Matrix Converters) supplies both the machines and
these machines are connected in parallel with appropriate phase transposition. The control
decoupling is possible due to decoupling of the α-β and x-y components. The d-q components
of one machine become the x-y to the other and vice-versa. Here, the vector controller
produces voltage references in contrary to series-connected drive where current references
are generated from the vector controller. Once again independent control is achieved of the
159
two five-phase motors. The voltage and current relations for this drive topology are given as
shown in equation (6.39);
21
21
21
21
21
desE
bdsD
ecsC
cbsB
aasA
vvv
vvv
vvv
vvv
vvv
21
21
21
21
21
desE
bdsD
ecsC
cbsB
aasA
iii
iii
iii
iii
iii
(6.39)
Fig. 6.5. Five-phase series-connected two-motor drive structure.
Fig. 6.6. Five-phase parallel-connected two-motor drive structure.
If current control in the rotating reference frame is to be utilized for vector control operation,
appropriate PWM scheme for five-phase supply source needs to be developed to generate
voltage references instead of current references for both series and parallel connected drives.
This proposes alternative solution of supplying the two-motor drive system using a three-
phase to five-phase Matrix Converter. The PWM technique develops and produces two-
frequency voltage reference outputs to control independently the two series/parallel
connected machines.
160
6.4b Carrier Based PWM Technique for a two motor system
Carrier-based PWM scheme developed in this section follows the similar concept presented
in section 6.1. Since the input side is three-phase, the analytical treatment remains the same
as that of single-motor drive system. However, the output is now increased to five and hence
the analysis will be modified to suit the requisite output phase number. A balanced three-
phase system is assumed at the input side.
)3/4tcos(Vv
)3/2tcos(Vv
)tcos(Vv
b
b
a
(6.40)
From equation (6.35) we can find that the generalised equation for output voltage of phase a
as:
)cos(Vk2
3V AA (6.41)
In equation (6.41), cos (ρ ) term indicates that the output voltage is affected by ρ . Thus, the
output voltage VA is independent of the input frequency and only depends on the amplitude V
of the input voltage and kA is a reference output voltage time-varying modulating signal for
the output phase A with the desired output frequency ω01 is the operating frequency of
machine-1 or the first fundamental output frequency and ω02 is the operating frequency of
machine-2 or the second fundamental output frequency. The fundamental output voltage
magnitude corresponding to ω01 is given as m1 and corresponding to ω02 is given as m2. The
five-phase reference output voltages can then be represented as:
1 1 01
1 1 01
1 1 01
1 1 01
1 1 01
cos( ),
cos( 2 / 5),
cos( 4 / 5),
cos( 6 / 5),
cos( 8 / 3),
A
B
C
D
E
k m t
k m t
k m t
k m t
k m t
(6.42)
1 2
1 2
1 2
1 2
1 2
A A A
B B B
C C C
D D D
E E E
K K K
K K K
K K K
K K K
K K K
(6.43)
161
Therefore the output voltages are obtained as:
1 01 2 02
1 01 2 02
1 01 2 02
1
3 3cos( ) cos( ) cos( ) cos( )
2 2
3 3 4cos( ) cos( 2 ) cos( ) cos( )
2 5 2 5
3 3cos( ) cos( 4 ) cos( ) cos( 8 )
2 3 2 5
3c
2
A A A
B B B
C C C
D D
V k V t k V t
V k V t k V t
V k V t k V t
V k V
01 2 02
1 01 2 02
3os( ) cos( 6 ) cos( ) cos( 2 )
5 2 5
3 3cos( ) cos( 8 ) cos( ) cos( 6 )
2 5 2 5
D
E E E
t k V t
V k V t k V t
(6.44)
The five phase reference output voltage can be written as:
1 01 2 02
1 1 01 2 02
1 1 01 2 02
1 1 01 2 02
1 1 01 2 02
cos( ) cos( ),
cos( 2 / 5) cos( 4 / 5)
cos( 4 / 5) cos( 8 / 5)
cos( 6 / 5) cos( 6 / 5)
cos( 8 / 3) cos( 2 / 5)
A
B
C
D
E
k m t m t
k m t m t
k m t m t
k m t m t
k m t m t
(6.45)
where ω is input frequency and 1o and 2o are the output frequencies of machine-1 and
machine-2 respectively, m1 and m2are the modulation indices for machine-1 and machine 2,
respectively. For the unity power factor ρ has to be chosen as zero. The modulating signals
are shown in Fig. 6.7, after adding the output common mode voltages. These duty ratios is
given in equation (6.34) are then compared with the high frequency triangular carrier signals
to generate the gating signals as illustrated in Fig. 6.7, for phase A. Similarly fifteen more
duty ratios will be compared with the triangular carrier to generate overall gating signals.
A complete block schematic of the PWM signal generation is presented in Fig. 6.8. The
reference voltages for two machines with the desired speeds and appropriate voltage
magnitudes are generated. These references are then summed according to the phase
transposition rule. The overall modulating signal thus generated is given to the PWM block.
This PWM block then generate appropriate gate signals for the Matrix Converter. The Matrix
Converter then produces appropriate voltages which drive the two series/ parallel connected
machines.
162
Fig. 6.7. Gate signal generation for output phase A.
Fig. 6.8. Block diagram of Carrier-based PWM for two frequency output.
6.4c Simulation Results for RL Load
MATLAB/SIMULINK model is developed for the proposed Matrix Converter control. The
input voltage is fixed at 100 V to show the exact gain at the output side. The switching
frequency of the devices is kept at 6 kHz. The purpose here is to show two fundamental
components of current produced by the Matrix Converter. These voltage components are
independent from each other and thus can independently control the two machines. The
results shown here is only limited to the production of the appropriate voltage components.
The motor behaviour is not discussed in this section. It is assumed that one voltage
component has frequency of 30 Hz and second voltage component has 60 Hz. To respect the
v/f=constant control the voltage magnitude of the lower frequency component is half
163
compared to the higher frequency component. For the simulation purpose, a R-L load is
connected with R = 10 Ω and L = 10 mH. Simulation results are shown for the modulation
with common mode voltage addition in the output target voltage. Thus the maximum output
of the Matrix Converter is limited to 78.8 V as the input is 100 V. The results without
common mode voltage addition will remain the same except with the lower output
magnitude. The resulting waveforms are presented in Figs. 6.9 - 6.10. The input source side
and converter side waveforms are presented in Fig. 6.9. The results clearly shows unity
power factor at the input side. The converter side current shows PWM signal and the
spectrum is clearly sinusoidal with no lower order harmonics.
Fig. 6.9. Input side waveforms of 3 to 5-phase Matrix Converter: upper trace.
164
Fig. 6.10. Input voltage and current, bottom trace, Spectrum Input current.
The output side filtered voltages are given in Fig. 6.11 and the spectrum of the output PWM
signal voltages are presented in Figs.6.11 - 6.13. The output filtered phase voltages shows
superimposed fundamental and second harmonic components. The spectrum of phase ‘A’
voltage and the transformed voltages are given in Fig.6.14. The phase voltage is transformed
into the two orthogonal planes and the voltage of α-axis and x-axis and their spectrum is
presented in Fig. 6.13 and Fig. 6.14, respectively. It is clearly evident that the phase ‘A’
voltage contains two fundamental components at 30 Hz and 60 Hz. These voltages are then
decoupled and appear in α-β plane (60 Hz) and x-y plane (30 Hz). Thus the aim of control is
achieved. Also the magnitude of the two voltages follows v/f=constant rule.
Fig. 6.11. Output filtered five-phase voltages.
165
Fig. 6.12. Spectrum of output voltages; phase ‘A’.
Fig. 6.13. Spectrum of output voltages; α-axis voltage.
166
Fig. 6.14. Spectrum of output voltages; x-axis voltage.
6.4d Simulation Results for Motor Load
The simulation model is developed in MATLAB/SIMULINK for the whole drive system.
Three-phase grid supply is assumed as 50 Hz, 440 V rms phase voltage (double voltage is
assumed since two-motor drive is considered). Five-phase reference voltage is chosen for the
first motor and another set of five-phase reference is assumed for the second motor. The five-
phase modulating signal is formulated by adding the two five-phase references according to
the transposition rule (Fig. 6.5). The parameter of the simulation is given in Table 6.1.
The simulation condition is taken as:
Motor-1 operating at rated speed of 1500 rpm (reference frequency of 50 Hz)
Motor-2 operating at half rated speed of 750 rpm (reference frequency of 5 Hz)
Load (half rated) applied to motor-1 at t = 1.2 sec
Load (one quarter of rated value) applied to motor-2 at t = 1.1
Switching frequency of the Matrix Converter is kept at 6 kHz.
The resulting waveforms for motor side and Matrix Converter sides are shown in Figs. 6.15
and 6.16, respectively.
167
Table 6.1: Simulation Parameters.
Parameters Name Parameters Values Source side resistance Rs 0.05 Ω Source side inductance Ls 8 mH DC link capacitor 2000 µF Stator resistance Rotor Resistance
10 Ω 6.5 Ω
Stator leakage inductance 40 mH Mutual inductance 420 mH Inertia J 0.03 kg sq m Number of Poles 4 Rated Torque Rated Voltage
8.33 Nm 440 V
a.
b.
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20
500
1000
1500
2000
Time [s]
Spee
ds M
otor
1 a
nd M
otor
2 [
rpm
] Motor1
Motor 2
-10
0
10
20
Tor
que
Mot
or 1
[N
m]
Motor2
Motor1
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
-20
0
20
40
Time [s]
TM
t2
[N]
168
c.
Fig. 6.15. Response of two-motor drive, a. Speeds, b. torques, c. phase ‘a’ current from Matrix Converter.
Fig. 6.16. Source side voltage and current, voltage is reduced to 150 times.
The spectrum for the output side current and voltage is shown in Figs. 6.17i and 6.17ii and
that of source side current filtered is shown in Fig. 6.17ii .
The output voltage and current waveform shows two frequency components at the two
operating frequencies of the two motors. The two machines show acceleration at the initial
response. When load is applied the speeds drops and the motor settles at the same speed. The
speed is not corrected as no closed-loop controller is employed in this analysis. The motor
torque is typical of a five-phase induction machine.
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-20
-15
-10
-5
0
5
10
15
20
Time [S]
Phas
e 'A
' MC
cur
rent
[A
]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1-3
-2
-1
0
1
2
3
Time [s]
Sour
ce s
ide
curr
ent a
ndV
olta
ge 1
50:1
[A
, V]
Voltage
Current
169
a.
b.
Fig. 6.17i. Matrix Converter output current and voltage time domain and frequency domain waveform.
Fig. 6.17ii. Filtered source side current spectrum.
The source side current is sinusoidal and working at unity power factor. This is the distinct
feature of the Matrix Converter based drives. The total harmonic distortion (THD) is
computed for the voltage and current as follows:
1.92 1.93 1.94 1.95 1.96 1.97 1.98 1.99 2-10
-5
0
5
Cur
rent
pha
se 'a
' [A
]
Time [s]
0 50 100 150 200 250 300 350 400 450 5000
2
4Fundamental = 3.946
Cur
rent
Spe
ctru
m [
A]
Frequency [Hz]
1.92 1.93 1.94 1.95 1.96 1.97 1.98 1.99 2-2000
0
2000
MC
Vol
tage
pha
se 'a
' [V
]
Time [s]
0 50 100 150 200 250 300 350 400 450 5000
200
400
Fundamental = 313.3556
Spec
trum
pha
se 'a
' vol
tage
[V
]
Frequency [Hz]
1.94 1.95 1.96 1.97 1.98 1.99 2-1
0
1
Sour
ce c
urre
nt [
A]
Time [s]
0 50 100 150 200 250 300 350 400 450 5000
0.2
0.4 Fundamental = 0.42673
Spec
trum
sou
rce
curr
ent [
A]
Frequency (Hz)
170
THD =
..7,5,3
2
1n
n
v
v (6.46)
where vn is the nth harmonic component and v1 is the fundamental component magnitude. For
the computation of THD, upto 10th lower order harmonic components are taken.
The THD for the source side current is calculated as 1.66%, the output side current THD is
4.55% while the output voltage THD is 4.48%. These values are well within the specified
limit.
6.5 Carrier Based PWM Technique for a Three-to-Six Phase Matrix Converter
Three to six-phase Matrix Converters consists of six legs with three bi-directional power
semi-conductor switches in series, which is capable to block voltage in both direction and
switch at high speed. Such system can be used to feed a quasi double-star six-phase drives.
Fig 6.18 illustrated the design of two step commutation scheme with a total number of 18
IGBTs. LC filter is used in input side to reduce the harmonics and output having two sets of
three phase outputs having 30o phase differences. Its inputs are the phase voltages Va, Vb, VC
and output voltages are VA ,VB, VC, VD, VE, VF. Eighteen bi-directional switches are
represented as S11 to S36.
sai
sbi
sci
ai
bi
ci
av
bv
cv
11S
21S
31S
12S
22S
32S
13S
23S
33S
14S
24S
34S
15S
25S
35S
16S
26S
36S
Ai Bi Ci Di Ei Fi
Fig. 6.18. Power Circuit topology of Three-phase to quasi six-phase Matrix Converter.
The load to the Matrix Converter is assumed as double star-connected six-phase ac machine
as shown in Fig.6-19.
171
A
B
C
D
E F
6/32 /
Fig. 6.19. Stator winding displacement of a quasi or double star six-phase ac machine.
6.5a Three-to-Six Phase Matrix Converter system
The balanced three-phase system is assumed at the input
)3/4tcos(Vv
)3/2tcos(Vv
)tcos(Vv
b
b
a
(6.47)
Since the Matrix Converter output voltages with frequency decoupled from the input
voltages, the duty ratios of the switches are to be calculated accordingly. The six-phase
output voltage duty ratios should be calculated in such a way that output voltages remains
independent of input frequency. In other words, the six-phase output voltages can be
considered in synchronous reference frame and the three-phase input voltages can be
considered to be in stationary reference frame, so that the input frequency term will be absent
in output voltages. Considering the above, duty ratios of output phase k is chosen as:
)/tcos(k
),/tcos(k
),tcos(k
kck
kbk
kak
34
32
(6.48)
where is the input displacement angle. Therefore the phase A output voltage can be
obtained by using the above duty ratios as
)]3/4tcos()3/4tcos()3/2tcos()3/2tcos()tcos()t[cos(VkV AA
(6.49)
The quasi-6-phase reference output voltages can be represented as:
172
),2/3tcos(mk
),3/4tcos(mk
),6/5tcos(mk
),3/2tcos(mk
),6/tcos(mk
),tcos(mk
oF
oE
oD
oC
oB
oA
(6.50)
Therefore, from equation (6.41), the output voltages are obtained as:
)2
3cos()cos(2
3
)3
4cos()cos(2
3
)6
5cos()cos(2
3
)3
2cos()cos(2
3
)6
2cos()cos(2
3
)cos()cos(2
3
tVkV
tVkV
tVkV
tVkV
tVkV
tVkV
oFF
oEE
oDD
oCC
oBB
oAA
(6.51)
6.5b Application of Offset Duty Ratio
The final modified duty ratios are shown in Fig. 6.20. If FEDCBA k,k,k,k,k,k are chosen to be
6-phase sinusoidal references as given in equation (9), the input voltage capability is not fully
utilized for output voltage generation and the output magnitude remains only 50% of the
input magnitude. To overcome this, an additional common mode term equal to
]2/)k,k,k,k,k,kmin()k,k,k,k,k,kmax( FEDCBAFEDCBA is added as in the carrier-based PWM
principle as implemented in two-level inverters. Thus the amplitude of )k,k,k,k,k,k( FEDCBA
can be enhanced from 0.5 to 0.5176.
173
Fig. 6.20. Modified offset duty ratios for all input phases.
6.5c With Harmonic Injection
The duty ratios can further be modified by harmonic injection of the output voltage
references to improve the output voltage magnitude. The output voltage magnitude increases
and reaches its limiting value of 88.6% of the input magnitude. The amount of harmonic
injection that is added to obtain new duty ratios are:
)6cos(6
10tmVcm (6.52)
The duty ratio for output phase p can be written as:
)3/4cos( ][3/))()()(1()(
)3/2cos( ][3/))()()(1()(
)cos(][3/))()()(1()(
tVktDtDtDtDd
tVktDtDtDtDd
tVktDtDtDtDd
cmpcbaccp
cmpcbabbp
cmpcbaaap
(6.53)
where F,E,D,C,B,Ap
The quasi six-phase output voltages can be written as:
),2/3cos(
),3/4cos(
),6/5cos(
),3/2cos(
),6/cos(
),cos(
tmk
tmk
tmk
tmk
tmk
tmk
oF
oE
oD
oC
oB
oA
(6.54)
where is input frequency and o is the output frequency, m is the modulation index. For
unity power factor has to be chosen as zero. The modulating signals are shown in Fig. 6.21,
after adding the output common mode voltages. The duty ratios obtained using equation
0 0.01 0.02 0.03 0.04 0.050
0.2
0.4
0.6
Time
Mag
nitu
de
OFFSET
174
(6.53) for phase A is depicted in Fig. 6.22. These duty ratios are then compared with the high
frequency triangular carrier signals to generate the gating signals as illustrated in Fig. 6.23,
for phase A. Similarly fifteen more duty ratios are compared with the triangular carrier to
generate overall gating signals.
Fig. 6.21. Common mode added reference for output phases.
Fig. 6.22. Duty ratio for output phase A.
Fig. 6.23.. Gate signal generation for output phase A.
0 0.01 0.02 0.03 0.04
-0.4
-0.2
0
0.2
0.4
0.6
Time (s)
Mag
nitu
de (
p.u.
)
0 0.01 0.02 0.03 0.040
0.5
1
0
0.5
1
0
0.5
1
Time (s)
0.0178 0.018 0.0182 0.0184
0
10
10
10
1
Time (s)
daA
dbA
dcA
S11
S21
S31
Carrier Signal
175
6.5d Output Voltage Magnitude
In a conventional three-phase to three-phase Matrix Converter, the output voltage magnitude
reaches 50% of the input voltage using the conventional modulation method or simple carrier
based PWM. This limit improves to 75% when the offset corresponding to the input signals
are added. It is important to note that this limit of 75% is irrespective of the output number of
phases of the Matrix Converter. Hence, this is the realizable output in Matrix Converter
topology with three phase input and n phase output. The output voltages are further enhanced
to 86.6% in three to three-phase Matrix Converter by adding common mode signal
corresponding to the output voltages. This increase in the output voltage magnitude is (86.6-
75)/75 = 15.47%, this is interesting value as the same increase is achieved in a two level
three-phase voltage source inverter when modulating signals are modified by adding common
mode voltages. Following the same principle the increase in the output voltage magnitude in
a multi-phase two-level inverter is obtained as n2/cos/1 , where n is the number of phases
of inverter given in reference [6.6]. This study of quasi six-phase Matrix Converter also
follows the same principle and an increase of (77.64-75)/75 = 3.52% in recorded. It is
important to mention here that the gain in the output voltage reduces with the increase in the
number of output phases of the Matrix Converter.
6.5e Simulation Results
MATLAB/SIMULINK model is developed for the proposed Matrix Converter control. The
input voltage is fixed at 100 V to show the exact gain at the output side. The switching
frequency of the devices is kept at 6 kHz. The output fundamental frequency is chosen as 25
Hz. Simulation results are shown for the modulation without common mode voltage addition
in the output target voltage. The load chosen is a quasi six-phase R-L with two separate
neutrals with R = 10 Ω and L = 10 mH. The results with common mode voltage addition will
remain the same except with the enhanced output magnitude. The resulting waveforms are
depicted in Figs. 6.24 and Fig. 6.25, for input side and output side, respectively. The results
of Fig. 6.25, clearly shows unity power factor at the input side and completely sinusoidal
source current. The input converter side current is seen as PWM signals due to switching and
current modulation. The spectrum of the input converter side current is also shown in the
lower trace. It is evident that the current is completely sinusoidal and does not contain any
low order harmonic. The total harmonic distortion in the input converter side current is very
low and is 0.84%. The output filtered voltages are seen as sinusoidal and the spectrum of the
output phase A voltage also shows the same results. The total harmonic distortion in the
176
output phase is observed as 0.95%. The load current is also completely sinusoidal. This
indicates the viability of the proposed modulation technique.
Fig. 6.24. Input side waveforms of 3 to quasi 6-phase Matrix Converter: upper trace, Input voltage and filtered current, bottom trace, Spectrum Input current.
(a)
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-100
-50
0
50
100
Time (s)
Inpu
t Pha
se 'a
' Vol
tage
& C
urre
nt (
V,A
)
0 0.02 0.04 0.06 0.08-20
0
20
Inpu
t Cur
rent
(A
)
Time (s)
10 50 100 1,000 10,0000
5
10
Fundamental = 11.63 A, THD=0.84%
Cur
rent
Spe
ctru
m (
A)
Frequency (Hz)
0 0.02 0.04 0.06 0.08-100
-80
-50
0
50
80
100
Time (s)
Filte
red
Out
put P
hase
Vol
tage
s (V
)
177
(b)
Fig. 6.25. Output side waveforms of 3 to quasi 6-phase Matrix Converter:, a) quasi 6-phase filtered output phase voltages, , b) Spectrum output voltage and output current.
6.6 Carrier Based PWM Technique for a Three-to-Six Phase Matrix Converter for Supplying Six-phase Two-motor Drives
As discussed in Chapter 4, vector control of any multi-phase machine requires only two stator
current components; the additional stator current components are used to control other
machines. It has been shown that, by connecting multi-phase stator windings in series and in
parallel with an appropriate phase transposition, it is possible to control independently all the
machines with supply coming from a single multi-phase voltage source inverter. Another
specific drive system, covered by this general concept, is the six-phase series-connected two-
motor drive, consisting of one symmetrical six-phase and a three-phase machine and supplied
from a single six-phase voltage source inverter. The multi-motor drive system discussed uses
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-10
-5
0
5
10
Time (s)
Out
put P
hase
Cur
rent
(A
)
0 0.02 0.04 0.06 0.08
-200
0
200
Out
put V
olta
ge (
V)
Time (s)
101
102
103
1040
50
100 Fundamental = 86.6, THD=0.95%
Vol
tage
Spe
ctru
m (
V)
Frequency (Hz)
178
multi-phase voltage source inverter as their supply. In contrast this section proposes a multi-
phase Matrix Converter to supply such drive topology.
In this section, a carrier based PWM strategy is presented based on the comparison of the
modulating signals (six-phase target voltages) with the high frequency triangular carrier wave
for a three to six-phase Matrix Converter. The output voltage is limited to 0.75 of the input
voltage magnitude. Theoretically this is the maximum output magnitude that can be obtained
in this Matrix Converter configuration in the linear modulation region. Analytical approach is
used to develop and analyse the proposed modulation techniques and are further supported by
simulation results. The major aim of this section is to produce two fundamental frequency
output from the Matrix Converter that can be used to control two series-connected six-phase
and three-phase machines.
6.6a Six-Phase Series-Connected Two-Motor Drive Configuration
The two-motor drive, used in this thesis, is shown in Figure 6.26. It consists of a six-phase
source (capital letters A, B,…F), a six-phase and a three-phase ac machine. Stator windings
of the two machines are connected in series with appropriate phase transposition. The six-
phase machine has the spatial displacement between any two consecutive stator phases of 60
(i.e. = 2/6). The type of the ac machine is irrelevant as long as the mmf distribution in the
air-gap is sinusoidal. Both machines are considered here as induction motors. Spatial
displacement between any two consecutive phases of the three-phase machine is 2 = 120.
The control of the two motors is decoupled although they are supplied from a single six-
phase source. The control decoupling is obtained by using indirect rotor field oriented
control. Flux and torque of the six-phase machine are controllable by source d-q axis current
components, while flux and torque of the three-phase machine can be controlled using source
x-y current components. The detailed modeling and experimental results of this configuration
are reported in reference [6.7], using current control in stationary reference frame. However,
current control in rotating reference frame requires an appropriate PWM for six-phase source.
This thesis presents as simple carrier-based PWM method for the three to six-phase Matrix
Converters which has not reported so far.
179
A
B C
D E F
Six-phase Source
as1 bs1 cs1 ds1 es1 fs1
as2 bs2 cs2
Machine 1: Six-phase IM
Machine 2: Three-phase IM
Fig. 6.26. Six-phase series-connected two-motor drive system.
6.6b Carrier-Based PWM Technique For Six-Phase Two-Motor Drive
The input and output voltages are related as:
c
b
a
cF
cE
bF
bE
aF
aE
cDbDaD
cCbCaC
cBbBaB
cAbAaA
F
E
D
C
B
A
V
V
V
V
VV
V
V
V
(6.56)
Therefore, the phase A output voltage can be obtained by using the duty ratio matrix
)]3/4cos()3/4cos()3/2cos(
)3/2cos()cos()[cos(
ttt
tttVkV AA
(6.57)
In equation (6.41), )cos( term indicates that the output voltage is affected by . The term kA
is defined in equation (6.59). Thus, the output voltage AV is independent of the input
frequency and only depends on the amplitude V of the input voltage and Ak is a reference
output voltage time-varying modulating signal for the output phase A with the desired output
frequency 21 oo , 1o is the operating frequency of machine-1 or the first fundamental
output frequency and 2o is the operating frequency of machine-2 or the second fundamental
output frequency. The fundamental output voltage magnitude corresponding to 1o is given
as m1 and corresponding to 2o is given as m2.The six-phase reference output voltages can
then be represented as:
180
6/10cos
)6/8cos(
)6/6cos(
)6/4cos(
)6/2cos(
),cos(
111
111
111
111
111
111
tmk
tmk
tmk
tmk
tmk
tmk
oF
oE
oD
oC
oB
oA
(6.58)
)3/4cos(
)3/2cos(
)cos(
222
222
222
tmk
tmk
tmk
oC
oB
oA
(6.59)
6.6c Application of Offset Duty Ratio
For the switches connected to output phase-A, at any instant, the condition 1d,d,d0 cAbAaA
should be valid. Therefore, offset duty ratios should to be added to the existing duty-ratios, so
that the net resultant duty-ratios of individual switches are always positive. Furthermore, the
offset duty-ratios should be added equally to all the output phases to ensure that the effect of
resultant output voltage vector produced by the offset duty-ratios is null in the load. That is,
the offset duty-ratios can only add the common-mode voltages in the output. Considering the
case of output phase-A:
21
21
21
21
21
21
2
12
12
12
12
12
1
CFF
BEE
ADD
CCC
BBB
AAA
kkk
kkk
kkk
kkk
kkk
kkk
(6.60)
Therefore, from equation (6.60), the output voltages are obtained as:
181
)cos(VkV
)cos(VkV
)cos(VkV
)cos(VkV
)cos(VkV
)cos(VkV
FF
EE
DD
CC
BB
AA
2
3
2
3
2
3
2
3
2
3
2
3
(6.61)
cos( ) cos( 2 / 3 ) cos( 4 / 3 ) 0aA bA cA A A Ak t k t k t
(6.62)
Absolute values of the duty-ratios are added to cancel the negative components from
individual duty ratios. Thus the minimum individual offset duty ratios should be:
)3/4cos()(
)3/2cos()(
)cos()(
tktD
and
tktD
tktD
AcAc
AbAb
AaAa
(6.63)
The effective duty ratios are:
)t(D
),t(D
),t(D
ccA'cA
bbA'bA
aaA'aA
(6.64)
Other output phases can be written in the similar fashion. The net duty ratio 10 'ak should
be within the range of 0 to 1.
For the worst case:
(6.65)
The maximum value of Ak is equal to 0.5. The duty-ratios are calculated for the other five
output phases. In the section (6.6b), two offsets are added to the original duty ratios to form
the following effective duty ratio for output phase p:
)3/4cos(3/))()()(1()(
)3/2cos(3/))()()(1()(
)cos(3/))()()(1()(
tktDtDtDtD
tktDtDtDtD
tktDtDtDtD
pcbaccp
pcbabbp
pcbaaap
(6.66)
120 Ak.
182
where F,E,D,C,B,Ap
These duty ratios (equation 6.66) can be compared to the triangular carrier wave to generate
the gating signals for the bi-directional power switches as shown in fig.6.27. The output
phase voltage magnitude will reach 75% of the input voltage magnitude with this method.
A complete block schematic of the PWM signal generation is presented in Fig. 6.28. The
reference voltages for two machines with the desired speeds and appropriate voltage
magnitudes are generated. These references are then summed according to the phase
transposition rule. The overall modulating signal thus generated is given to the PWM block.
This PWM block then generate appropriate gate signals for the Matrix Converter. The Matrix
Converter then produces appropriate voltages which drive the two series/parallel connected
machines.
Fig. 6.27. Gate signal generation for output phase A.
PWM
AddAsPer
Transpose Rule
Ref-6-phase
Ref-3-phase Carrier
3-phase Input
6-phaseOutput
GateDrive
Modulating Signal
Fig. 6.28. Block diagram of Carrier-based PWM for two frequency output.
6.6d Simulation Results
MATLAB/SIMULINK model is developed for the proposed Matrix Converter control. The
input voltage is fixed at 100 V to show the exact gain at the output side. The switching
0.0178 0.018 0.0182 0.0184
0
10
10
10
1
Time (s)
daA
dbA
dcA
S11
S21
S31
Carrier Signal
183
frequency of the devices is kept at 6 kHz. The purpose here is to show two fundamental
components of current produced by the Matrix Converter. These voltage components are
independent from each other and thus can independently control the two machines.
The results shown in this thesis is only limited to the production of the appropriate voltage
components. The motor behaviour is not discussed in this thesis and will be reported
separately. It is assumed that one voltage component has frequency of 25 Hz and second
voltage component has 50 Hz. To respect the v/f=constant control the voltage magnitude of
the lower frequency component is half compared to the higher frequency component. For the
simulation purpose a R-L load is connected with R = 10 Ω and L = 10 mH. Simulation results
are shown for the modulation with common mode voltage addition in the output target
voltage. Thus the maximum output of the Matrix Converter is limited to 75 V as the input is
100 V.
The resulting waveforms are presented in Figs. 6.29 to 6.31. The input source side and
converter side waveforms are presented in Fig. 6.29. The results clearly shows unity power
factor at the input side. The converter side current shows PWM signal and the spectrum is
clearly sinusoidal with no lower order harmonics. The output side filtered voltages are given
in Fig. 6.30 and one phase filtered voltage and current is illustrated in Fig. 6.30. The
spectrum of the output PWM signal voltages are presented in Fig. 6.31. The output filtered
phase voltages shows superimposed fundamental and second harmonic components. The
spectrum of phase ‘A’ voltage and the transformed voltage are given in Fig. 6.31. It is clearly
evident that the phase ‘A’ voltage contains two fundamental components at 25 Hz and 50 Hz.
These voltages are then decoupled and appear in α-β plane (50 Hz) and x-y plane (25 Hz).
Thus is aim of the control is achieved. Also the magnitude of the two voltages follows
v/f=constant rule.
184
a.
b.
Fig. 6.29 Input side waveforms of 3 to 5-phase Matrix Converter: (a) Input voltage and current, (b) Spectrum Input current.
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04-100
-50
0
50
100
Time [s]
Sou
rce
side
vol
tage
and
cur
rent
pha
se 'a
' [V
, A]
Voltage
Current
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-10
-5
0
5
10
Selected signal: 4 cycles. FFT window (in red): 2 cycles
Time [s]
Sou
rce
curr
ent [
A]
0 200 400 600 800 10000
50
100
Frequency [Hz]
Fundamental (50Hz) = 6.64 A
Spe
ctru
m [
A]
185
a.
b.
Fig. 6.30. Output waveforms, a. filtered voltages, b. phase ‘a’ output filtered voltage and current.
a.
0.025 0.03 0.035 0.04 0.045 0.05 0.055 0.06 0.065 0.07 0.075-80
-60
-40
-20
0
20
40
60
80
Time [s]
Out
put
filt
ered
pha
se v
olta
ges
[V]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-80
-60
-40
-20
0
20
40
60
80
Time [s]
Out
put p
hase
'a' v
olta
ge a
nd c
urre
nt [
V,A
]
Current
Voltage
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-200
0
200Selected signal: 2 cycles. FFT window (in red): 2 cycles
Time [s]Pha
se 'a
' vol
tage
[V
]
0 200 400 600 800 10000
50
100
150
200
Frequency [Hz]
Fundamental (25Hz) = 24.66 V
Spe
ctru
m [
V]
186
b.
c. Fig. 6.23. Spectrum of output voltages; a. phase ‘A’, b. α-axis voltage and c. x-axis voltage.
6.7 Carrier Based PWM Technique Strategies for a Three-to-Seven Phase Matrix Converter
The power circuit topology of a three-phase to seven-phase Matrix Converter is illustrated in
Fig. 6.32. There are seven legs with each leg having three bi-directional power switches
connected in series. Each power switch is bi-directional in nature with anti-parallel connected
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-100
0
100
200Selected signal: 4 cycles. FFT window (in red): 2 cycles
Time [s]
-a
xis
Vol
tage
[V
]
0 200 400 600 800 10000
50
100
Frequency [Hz]
Fundamental (50Hz) = 48.76 V
Spe
ctru
m [
V]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-100
0
100
Selected signal: 2 cycles. FFT window (in red): 2 cycles
Time [s]
x-ax
is v
olta
ge [
V]
0 200 400 600 800 10000
50
100
Frequency [Hz]
Fundamental (25Hz) = 24.64 V
Spec
trum
[V
]
IGBTs
having
between
Fig. 6.3
The 7
6.7a Ap
In the e
the swi
should b
that the
offset d
resultan
the offs
case of
and diodes
LC filters
n each phas
32. Power C
7-phase refe
pplication o
equation (6.
itches conn
be valid. Th
e net resulta
duty-ratios s
nt output vo
set duty-rati
output phas
s. The inpu
and the o
ses.
Circuit topol
erence outpu
of Offset D
.67), duty-r
nected to ou
herefore, of
ant duty-rati
should be ad
oltage vecto
ios can only
se-A:
k
k
k
k
k
k
k
G
F
E
D
C
B
A
ut is similar
output is se
logy of Thre
ut voltages c
uty Ratio
ratios becom
utput phase
ffset duty ra
ios of indiv
dded equall
or produced
y add the co
cos(
cos(
cos(
cos(
cos(
cos(
cos(
tm
tm
tm
tm
tm
tm
tm
oG
oF
oE
oD
oC
oB
oA
187
r to a three
even phase
ee-phase to
can be repre
me negative
e-A, at any
atios should
vidual switch
ly to all the
d by the offs
ommon-mod
),7/12
),7/10
),7/8
),7/6
),7/4
),7/2
),
t
t
t
t
t
t
t
e-phase to
s with 51.4
seven-phas
esented as:
e which are
instant, th
to be added
hes are alw
output pha
set duty-rati
de voltages
three-phase
4 degree p
se Matrix Co
not practic
he condition
d to the exis
ways positiv
ases to ensu
ios is null i
in the outp
e Matrix Co
phase displa
onverter.
cally realiza
n
sting duty-r
e. Furtherm
ure that the e
in the load.
put. Conside
,0 bAaA dd
onverter
acement
(6.67)
able. For
ratios, so
more, the
effect of
That is,
ering the
1, cAA d
188
(6.69)
Absolute values of the duty-ratios are added to cancel the negative components from
individual duty ratios. Thus the minimum individual offset duty ratios should be:
( ) ( )
2( ) ( )
3
4( ) ( )
3
a A
b A
b A
D t K cos wt
D t K cos wt
D t K cos wt
(6.70)
The effective duty ratios are . Other output phases can be
written similarly. The net duty ratio should be accommodated within a range of 0
to 1. Therefore,
can be written as:
(6.71)
For the worst case
(6.72)
The maximum value of or in other words k in equation (6.73) is equal to 0.5 or sin(π/6).
Hence the offset duty-ratios corresponding to the three input phases are chosen as:
(6.73)
The modified duty ratios for output phase A are:
(6.74)
In any switching cycle the output phase should not be open circuited. Thus the sum of the
duty ratios in equation (6.74) must equal to unity. But the summation is
less than or equal to unity. Hence another offset duty-ratio is
0)3/4cos(
)3/2cos(
)cos(
tk
tk
tkddd
A
A
AcAbAaA
)(),(),( tDdtDdtDd ccAbbAaaA
)t(Dd aaA
1)(0 tDd aaA
1)cos()cos(0 tktk AA
Ak
)3/4cos(5.0)(
)3/2cos(5.0)(
,)cos(5.0)(
ttDand
ttD
ttD
c
b
a
)3/4cos()(
),3/2cos()(
),cos()(
tktDd
tktDd
tktDd
AccA
AbbA
AaaA
)t(D)t(D)t(D cba
31 /)t(D)t(D)t(D cba
2.20 Ak
189
added to . The addition of this offset duty-ratio in all switches will maintain
the output voltages and input currents are unaffected. The equation (6.74) derives the
maximum modulation index for three phase input with seven phase output from equation
(6.68) as .
If are chosen to be 7-phase sinusoidal references, the input voltage
capability is not fully utilized for output voltage generation. To overcome this, an additional
common mode term equal to is added as
in the carrier-based space-vector PWM principle as implemented in two-level inverters. Thus
the amplitude of can be enhanced from 0.5 with 0.5129.This is
shown in fig.6.33. for an output phase. Therefore the maximum modulation index that can be
achieved in the linear modulation range with common mode addition is
.
In the section (6.7b) the analytical expressions are given for the case of a three to seven phase
Matrix Converter.
6.7b Without Common-mode voltage addition
The duty ratio for output phase A can be written as:
(6.75)
6.7b With Common mode voltage Addition
The duty ratio for output phase A can be written as:
)t(D),t(D),t(D cba
75%or 75.0)6
sin(2
3
2
3
Ak
GFEDCBA kkkkkkk ,,,,,,
]2/),,,,,,min(),,,,,,max( GFEDCBAGFEDCBA kkkkkkkkkkkkkk
),,,,,,( GFEDCBA kkkkkkk
76.94%or 76940512902
3
2
3..kA
)/tcos(k
/))t(D)t(D)t(D()t(Dd
)/tcos(k
/))t(D)t(D)t(D()t(Dd
)tcos(k
/))t(D)t(D)t(D()t(Dd
A
cbaacA
A
cbaabA
A
cbaaaA
34
31
32
31
31
190
(6.76)
where
(6.77
The seven-phase output voltages can be written as:
(6.78)
where is input frequency and is the output frequency, m is the modulation index. For
unity power factor has to be chosen zero.
6.7c Simulation Results
MATALAB/SIMULINK model is developed for the proposed three to seven-phase Matrix
Converter control. The input voltage is fixed at 100 V to show the exact gain at the output
side. The switching frequency of the devices is kept at 6 kHz. The fundamental frequency of
the output is chosen as 25 Hz. Simulation results are shown for the modulation with common
mode voltage addition in the output target voltage. The resulting waveforms are presented in
Figs. 6.33 - 6.34. The results of source voltage and source current clearly shows unity power
factor at the source side, Fig. 6.33. The filtered output seven-phase voltages are shown in Fig.
6.33 and are seen to be as sinusoidal. The magnitude of the fundamental component proves
the enhanced output without any lower-order harmonics. This indicates the viability of the
)3/4cos(]2/),,,,,,min(
),,,,,,max([
3/))()()(1()(
)3/2cos(]2/),,,,,,min(
),,,,,,max([
3/))()()(1()(
)cos(]2/),,,,,,min(
),,,,,,max([
3/))()()(1()(
tkkkkkkk
kkkkkkkk
tDtDtDtDd
tkkkkkkk
kkkkkkkk
tDtDtDtDd
tkkkkkkk
kkkkkkkk
tDtDtDtDd
GFEDCBA
GFEDCBAA
cbaacA
GFEDCBA
GFEDCBAA
cbaabA
GFEDCBA
GFEDCBAA
cbaaaA
)3/4cos(5.0)(
)3/2cos(5.0)(
)cos(5.0)(
ttD
ttD
ttD
b
b
a
)7/12cos(
)7/10cos(
)7/8cos(
)7/6cos(
)7/4cos(
)7/2cos(
)cos(
tmk
tmk
tmk
tmk
tmk
tmk
tmk
oG
oF
oE
oD
oC
oB
oA
o
191
proposed modulation technique. Both simulation as well as experimental results are taken
using sample time of 50 microsecond.
Fig. 6.33. Input side waveforms of 3 to 7-phase Matrix Converter.
Fig. 6.34. Output side waveforms of 3 to 7-phase Matrix Converter, seven-phase filtered
output phase voltages.
The results with input side over modulation are presented in Fig. 6.35 and 6.36. The
distortion in the source side current is clearly visible in Fig. 6.35. Additionally the distortion
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-100
-80
-60
-40
-20
0
20
40
60
80
100
Time(sec)
Inpu
t V
olta
ge &
Cu
rren
t(V
,A)
Voltage
Current
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-20
0
20
Inp
ut
Cu
rren
t(A
)
Time (sec)
50 200 400 600 800 1,0000
5
10
15
Frequency(Hz)
Cu
rren
t S
pec
tru
m(A
)
0 0.02 0.04 0.06 0.08-100
-50
0
50
100
Time (sec)
Out
put f
ilter
ed P
hase
Vol
tage
s (V
)
192
in the load current is also evident from Fig. 6.35. The distortion in the voltage can be seen
form Fig. 6.36.
Fig. 6.35 Spectrum output voltage.
.
Fig. 6.36. Input side waveforms of 3 to 7-phase Matrix Converter: Voltage and current with input side over-modulation.
Fig. 6.37. Output side current waveforms of 3 to 7-phase Matrix Converter with input side
over-modulation.
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-200
0
200
Ou
tpu
t V
olta
ge(V
)Time (sec)
25 200 400 600 800 1,0000
25
50
75
100
Frequency(Hz)
Vol
tage
Sp
ectr
um
(V)
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-100
-50
0
50
100
Time (sec)
Inpu
t Vot
age
& C
urre
nt (
V,A
)
0 0.02 0.04 0.06 0.08-10
-5
0
5
10
Time (sec)
Out
put C
urre
nts
(A)
193
The THD output current with and without over modulation are 2.82% and 2%, respectively.
While the THD of the input current with and without over modulation are 7.53% and 1.65%,
respectively.
6.8 Seven-Phase Series Connected Three-Motor Drive System
The three-motor drive, used in the reference [6.8], is shown in Figure 6.38. Stator windings of
the three machines are connected in series with appropriate phase transposition. The seven-
phase machine has the spatial displacement between any two consecutive stator phases of
(i.e. = 2/7). It is to be noted here that the proposed PWM scheme is equally valid for
series connection as well as parallel connection; however, only series-connected drive is
elaborated here.
When the phase variable equations are transformed using decoupling matrix, four sets of
equations are obtained, namely d-q, x1-y1, x2-y2 and zero sequence. In single seven-phase
motor drive, the d-q components are involved in actual electromagnetic energy conversion
while the x1-y1 and x2-y2 components increase the thermal loading of the machine. However,
the extra set of current components (x1-y1 and x2-y2) available in a seven-phase system is
effectively utilised in independently controlling two additional seven-phase machines when
the stator windings of three seven-phase machines are connected in series and are supplied
from a single seven-phase VSI. Reference currents/voltages generated by three independent
vector controllers, are summed up as per the transposition rules and are supplied to the series-
connected connected seven-phase machines. Block diagram of the three-motor drive systems
is illustrated in Fig. 6.38 for series connection and the connectivity matrix is in Table 6.2.
Table 6.2. Connectivity matrix for the seven-phase series connected three-motor case.
A B C D E F G M1 1 2 3 4 5 6 7 M2 1 3 5 7 2 4 6 M3 1 4 7 3 6 2 5
194
Fig. 6.38. Seven-phase series-connected three-motor drive structure.
The scheme of series-connected seven-phase three-motor drive discussed in the reference
[6.8] utilizes current control in the stationary reference frame. However, if current control in
the rotating reference frame is to be utilized, appropriate PWM scheme for seven-phase VSI
needs to be developed to generate voltage references instead of current references.
6.8a Carrier-Based Pulse Width Modulation Technique
Carrier-based PWM scheme developed in this section follows the similar concept presented
in reference [6.5]. Since the Matrix Converter output voltages with frequency decoupled from
the input voltages, the duty ratios of the switches are to be calculated accordingly. The seven-
phase output voltage duty ratios should be calculated in such a way that output voltages
remain independent of input frequency. In a different way the seven-phase output voltages
can be considered in synchronous reference frame and the three-phase input voltages can be
considered to be in stationary reference frame, so that the input frequency term will be absent
in output voltages. Considering the above points duty ratios of output phase j where are
chosen as:
)3/4cos(
),3/2cos(
),cos(
tk
tk
tk
jck
jbk
jak
(6.79)
Therefore the phase A output voltage can be obtained by using the above duty ratios as:
)]3/4cos()3/4cos(
)3/2cos()3/2cos(
)cos()[cos(
tt
tt
ttVkv AA
(6.80)
195
The individual references of three machines are considered as:
),7/12cos(
),7/10cos(
),7/8cos(
),7/6cos(
),7/4cos(
),7/2cos(
),cos(
11
11
11
11
11
11
11
tmk
tmk
tmk
tmk
tmk
tmk
tmk
oG
oF
oE
oD
oC
oB
oA
(6.81)
),7/12cos(
),7/10cos(
),7/8cos(
),7/6cos(
),7/4cos(
),7/2cos(
),cos(
22
22
22
22
22
22
22
tmk
tmk
tmk
tmk
tmk
tmk
tmk
oG
oF
oE
oD
oC
oB
oA
(6.82)
),7/12cos(
),7/10cos(
),7/8cos(
),7/6cos(
),7/4cos(
),7/2cos(
),cos(
33
33
33
33
33
33
33
tmk
tmk
tmk
tmk
tmk
tmk
tmk
oG
oF
oE
oD
oC
oB
oA
(6.83)
Since the three machines are connected in series with phase transposition, the references
should also be created accordingly as given in the following equation:
321
321
321
321
321
321
321
EFGG
BDFF
FBEE
CGDD
GECC
DCBB
AAAA
kkkk
kkkk
kkkk
kkkk
kkkk
kkkk
kkkk
(6.84)
The output voltage in phase-A is:
tttVkv oooAA 321 coscoscos)cos(
2
3
(6.85)
196
Similarly, the relationship for other six phases can be written. For the switches connected to
output phase-A, at any instant, the condition 1,,0 cAbAaA should be valid. Therefore,
offset duty ratios should to be added to the existing duty-ratios, so that the net resultant duty-
ratios of individual switches are always positive. Furthermore, the offset duty-ratios should
be added equally to all the output phases to ensure that the effect of resultant output voltage
vector produced by the offset duty-ratios is null in the load. That is, the offset duty-ratios can
only add the common-mode voltages in the output. Considering the case of output phase-A:
03
4cos
3
2coscos
tk
tktk
A
AAcAbAaA
(6.86)
Absolute values of the duty-ratios are added to cancel the negative components from
individual duty ratios. Thus the minimum individual offset duty ratios should be
( ) cos( ) , ( ) cos( 2 / 3 )
( ) cos( 4 / 3 )
a A b A
c A
D t k t D t k t
and D t k t
(6.87)
The effective duty ratios are )(),(),( tDtDtD ccAbbAaaA , Other output phases can be written
similarly. The net duty ratio )(tDaaA should be accommodated within a range of 0 to 1.
Therefore 1)(0 tDaaA can be written as 1)cos()cos(0 tktk AA
The maximum value of Ak is equal to 0.5. Hence the offset duty-ratios corresponding to the
three input phases are chosen as:
)3/4cos(5.0)(
)3/2cos(5.0)(
,)cos(5.0)(
ttDand
ttD
ttD
c
b
a
(6.88)
The modified duty ratios for output phase A are
)3/4cos()(
),3/2cos()(
),cos()(
tktD
tktD
tktD
AccA
AbbA
AaaA
(6.89)
In any switching cycle the output phase has to be connected to any of the input phases. The
summation of the duty ratios in equation (6.89) must equal to unity. But the summation
197
)()()( tDtDtD cba is less than or equal to unity. Hence another offset duty-ratio
3/)()()(1 tDtDtD cba is added to )( )( ),( tDandtDtD cba in equation (6.89). The addition of this
offset duty-ratio in all switches will maintain the output voltages and input currents
unaffected. Similarly, the duty-ratios are calculated for the other output phases.
If GFEDCBA kkkkkkk ,,,,,, are chosen to be 7-phase sinusoidal references as given in equation
(6.85), the input voltage capability is not fully utilized for output voltage generation. To
overcome this, an additional common mode term equal to
]2/),,,,,,min(),,,,,,max( GFEDCBAGFEDCBA kkkkkkkkkkkkkk is added as in the carrier-based space-vector
PWM principle as implemented in two-level inverters. Thus the amplitude of
),,,,,,( GFEDCBA kkkkkkk can be enhanced from 0.5 with 0.5129.
The duty ratio for output phase A can be written as:
)3/4cos(]2/),,,,,,min(
),,,,,,max([
3/))()()(1()(
)3/2cos(]2/),,,,,,min(
),,,,,,max([
3/))()()(1()(
)cos(]2/),,,,,,min(
),,,,,,max([
3/))()()(1()(
tkkkkkkk
kkkkkkkk
tDtDtDtD
tkkkkkkk
kkkkkkkk
tDtDtDtD
tkkkkkkk
kkkkkkkk
tDtDtDtD
GFEDCBA
GFEDCBAA
cbaacA
GFEDCBA
GFEDCBAA
cbaabA
GFEDCBA
GFEDCBAA
cbaaaA
(6.90)
where
)3/4cos(5.0)(
)3/2cos(5.0)(
)cos(5.0)(
ttD
ttD
ttD
b
b
a
(6.91)
6.8b Simulation Results
MATLAB/SIMULINK model is developed for the proposed Matrix Converter control. The
input voltage is fixed at 100 V to show the ratio of output to input voltage gain. The
switching frequency of the devices is chosen as 6 kHz. The purpose here is to show three
fundamental components of current/voltage produced by the seven-phase Matrix Converter.
These voltage components are independent from each other and thus can independently
control the three machines whose stator windings are connected in series. The results shown
here is only limited to the production of the appropriate voltage components. The motor
behavior is not discussed here. The objective of controlling three machines will be
198
accomplished if three independent voltages with different frequency components are
generated.
6.8c Independent Control at Identical Frequencies
It is assumed, that the three supplied machines operate at same speeds. To fulfil such
requirement, Matrix Converter needs to generate three independent frequency output in the
output voltage waveforms. To prove the concept of decoupled control, simple R-L load is
considered. The requirement imposed on the Matrix Converter, is the generation of three
independent frequency component in three orthogonal planes (i.e. in d-q, x1-y1 and x2-y2).
The fundamental frequency is chosen as 50 Hz for all the three motors. For the simulation
purpose a R-L load is connected with R = 10 Ω and L = 10 mH. Simulation results are shown
for the modulation with common mode voltage addition in the output target voltage. Thus the
maximum output of the Matrix Converter is limited to 75 V as the input is 100 V.
The resulting waveforms are presented in Figs. 6.39 to 6.41. The input source side voltage
and currents waveforms are presented in Fig. 6.39. The result waveforms show unity power
factor operation of the Matrix Converter. The output side filtered voltages are given in Fig.
6.40 The spectrum of the output PWM signal voltages are presented in Fig. 6.41. The
spectrum of phase ‘A’ voltage is clearly evident that the phase ‘A’ voltage contains three
fundamental components at 50 Hz (i.e. the sum of the three voltages at the same frequency).
These voltages are then transformed to show in d-q plane (50 Hz), x1-y1 plane (50 Hz) and
x2-y2 plane (50 Hz).
Fig. 6.39. Input side waveforms of 3 to 7-phase Matrix Converter.
199
Fig. 6.40. Output waveforms, filtered output voltages.
a.
b.
200
c.
d. Fig. 6.24. Spectrum of output voltages; a. phase ‘A’, b. α-axis voltage c. x1-axis voltage, d.
x2-axis voltage.
Two more performance indices for comparison purpose are considered namely Total
Harmonic Distortion (THD) and Weighted Total Harmonic Distortion (WTHD). The method
of comparing the effectiveness of modulation is by comparing the unwanted components i.e.
the distortion in the output voltage or current waveform, relative to that of an ideal sine wave,
it can be assumed that by proper control, the positive and negative portions of the output are
symmetrical (no DC or even harmonics).
Normalizing this expression of THD with respect to the quantity (V1) i.e. fundamental, the
weighted total harmonic distortion (WTHD) becomes defined as
201
1
..7,5,3
2
V
n
V
WTHDn
n
(6.92)
For d-axis output voltage having a single component at 50 Hz frequency of magnitude 24.96
V and has THD = 4.38% of the fundamental and WTHD = 2.46% of the fundamental. For
x1-axis output voltage having a single component at 50 Hz frequency of magnitude 24.96 V
and has THD = 4.30% of the fundamental and WTHD = 1.93% of the fundamental. For x2-
axis output voltage having a single component at 50 Hz frequency of magnitude 25.55Vand
has THD = 4.64% of the fundamental and WTHD = 2.63 % of the fundamental. It shows that
the output voltage is nearly equal to the sinusoidal value. Thus aim of the control is achieved.
6.8d Independent Control at Three Different Frequencies
It is assumed that one machine is running at rated speed and the second machine at half of the
rated and third at one fourth of the rated. One voltage component has frequency of 12.5 Hz,
second voltage component has frequency of 25 Hz and third voltage component has 50 Hz.
To respect the v/f=constant control the voltage magnitude of the lower frequency component
has compared to the higher frequency component with appropriate ratio.
The resulting waveforms are presented in Figs. 6.42 to 6.43. The spectrum of the output
PWM signal of phase ‘A’ voltage and the transformed voltage are given in Fig. 6.42. It is
clearly evident that the phase ‘A’ voltage contains three fundamental components at 12.5 Hz,
25 Hz and 50 Hz. These voltages are then decoupled and appear in α-β plane (12.5 Hz), x1-
y1 plane (50 Hz) and x2-y2 plane (25 Hz). This shows independent control or decoupled
control.
The Matrix Converter alpha-axis phase ‘A’ voltage in Fig. 6.42a. shows that it contains a
single component at 12.5 Hz frequency with a magnitude of 5.49 V and has THD = 4.80% of
the fundamental and WTHD = 4.10 % of the fundamental. Fig. 6.38(c) shows single
component at 50 Hz frequency with a magnitude of 28.26 V in x1-y1 axis and has THD = 1.84
% of the fundamental and WTHD = 1.36 % of the fundamental. and Fig. 6.38(d) shows
single component at 25 Hz frequency with a magnitude of 17.34 V in x2-y2 axis and has THD
= 4.31% of the fundamental and WTHD = 4.20 % of the fundamental.
202
a.
b.
c.
203
d.
Fig. 6.42. Spectrum of output voltages; a). phase ‘A’, b). α-axis voltage c). x1-axis voltage d). x2 - axis voltage.
Fig. 6.43 is showing the filtered output phase ‘a’ volatge of the seven phases after
connecteing RL load at the output terminals.
Fig. 6.25. Output waveforms, filtered output phase ‘a’ voltage.
6.9 Experimental Investigation
This section describes the experimental set-up that is developed for the validation of the
simulation results and also the experimental results are shown.
6.9a Experimental Set-up
204
A prototype three-phase to nine-phase Matrix Converter is developed where input is three-
phase and the output can be configured from single to nine phases. This is done in order to
develop a modular multi-phase Matrix Converter where variety of experiments can be
conducted. The set-up consists of hardware and software part. Hardware is related to the
power switching devices that are arranged in the 3 x 9 array with associated auxiliaries such
as input side filter and clamp circuit for protection. To control this Matrix Converter, control
platform based on DSP/FPGA combination is developed. Since the DSP solution alone
cannot be used for this application due to limited number of available PWM channels in a
DSP chip. Hence, a solution is developed that is based on the integration of DSP and FPGA
chip, where the processing is done in DSP core while the PWM is generated from FPGA.
This is the most suitable control platform that can be developed for such complex system.
The DSP/FPGA can be coded either in C++ using code compose studio compiler or in system
generator from Xilinx. The DSP/FPGA board is interfaced with the PC using JTag emulator.
The proposed carrier based modulation scheme and direct duty ratio based PWM are
implemented for three to five- phase and three-to-six phase in quasi configuration Matrix
Converter for single and two-frequency output. The results of carrier-based scheme are
shown in this chapter while the results of direct duty ratio based PWM is reported in Chapter
7.
The overall block schematic of the experimental set-up is presented in Fig. 6.44. The block
diagram of the power module (3 x 9 phase Matrix Converter) is presented in Fig. 6.45 and
detailed drawing is shown in Fig. 6.46. The pictorial views of different components are given
in Fig. 6.47.
The power module is bi-directional switch FIO 50-12BD from IXYS and is composed of a
diagonal IGBT and fast diode bridge in ISOPLUS i4-PACTMas shown in Fig. 6.48. The
voltage blocking capability of the device is 1200 V and the current capacity is 50A. It
controls bi-directional current flow by a single control signal. This comes in single chip with
five output pins; four for the diode bridge and one for the gate drive of the IGBT. The
advantage of this bi-directional power switch is the decreased number of IGBTs which is a
major issue for multi-phase operation, but the major disadvantage is the higher conduction
losses and the two-step commutation. The Matrix Converter consists of eighteen such bi-
directional power switches.
Extra line inductances are used for safe operation during the overlapping of current
commutation. Dead-time compensation is done along with snubbers and clamping circuit.
205
Input filter are used to reduce the switching frequency harmonics present in the input current.
This filter circuit frequency does not need to store energy coming from the load. Simple LC
filters used for this purpose.
Requirements of LC filter are
* Cut-off frequency lower than the switching frequency of Matrix Converter.
* Minimal reactive power at grid frequency.
* Minimal volume and weight.
* Minimal filter inductance voltage drop at rated current in order to avoid reduction in
voltage transfer ratio.
Careful filter design is necessary else it will affect both the input and output currents of the
converter. Block diagram of the input filter is shown in Fig. 6.50.
The control platform used is Spartan 3-A DSP controller and Xilinx XC3SD1800A FPGA as
shown in Fig.6.49. The modulation code is written in C++ and is processed in the DSP. The
logical tasks such as A/D and D/A conversion, gate drive signal generation etc are
accomplished by the powerful FPGA board. The FPGA board is able to handle upto 50 PWM
signals.
In a Matrix Converter, over voltages can appear from the input side, caused due to line
perturbations over voltages can also appear from output side caused due to an over current
fault. When switches are turned OFF, the current in the load is suddenly interrupted. The
energy stored in the motor inductance has to be discharged without creating over-voltages. A
clamp circuit is the common solution to avoid both input and output over voltages. This
clamp circuit has 24 fast-recovery diode to connect capacitor to input and output terminals.
Block diagram of the clamp circuit used in the set-up is presented in Fig. 6.52.
206
Fig. 6.44. Overall block schematic of the experimental set-up.
Fig. 6.45. Block schematic of experimental set-up of three-phase to Nine-phase Matrix Converter.
I-12
I-2 I-3 I-4 I-5
I-7 I-6
I-8
I-9
I-10 I-11
VPE Sparten 3A
DSP
FPGA CONTROLLER
Opto-Isolation&
Gate Driver
Matrix Converter
Power Module
I/P- Filter
3Phase
R Y B
Rectifier 1
Rectifier 2
+
-
+
-
Clamping Circuit
RL- Load
N
Front Panel Test Points
Current Transducer
1
23
26
27
1
23
26
27
Rheostat
C
Fig. 6.46. Detailed drawing of the experimental set-up.
207
Fig. 6.47. An experimental setup photo graph (main power circuit).
Fig. 6.48. Nine units of Matrix Converter.
208
Fig. 6.49. Spartan DSP 3A evaluation board.
Fig. 6.50. Bi-directional switching arrangement.
Fig. 6.51. Input filter structure.
209
Fig. 6.52. Clamp circuit.
At the input side, an isolation transformer is required. An isolation transformer is a
transformer, often with symmetrical windings, which is used to decouple two circuits. An
isolation transformer allows an AC signal or power to be taken from one device and fed into
another without electrically connecting the two circuits. Isolation transformers block
transmission of DC signals from one circuit to the other, but allow AC signals to pass. They
also block interference caused by ground loops. Isolation transformers with electrostatic
shields are used for power supplies for sensitive equipment such as computers or laboratory
instruments.
Isolation transformers are commonly designed with careful attention to capacitive coupling
between the two windings. This is necessary because excessive capacitance could also couple
AC current from the primary to the secondary. A grounded shield is commonly interposed
between the primary and the secondary. Any remaining capacitive coupling between the
secondary and ground simply causes the secondary to become balanced about the ground
potential.
Overall block diagram of the Matrix Converter connection for experimental investigation is
shown in Fig. 6.53. A Matrix Converter is implemented by a matrix module which consists
of 9 bi-directional switches. The 3-phase input supply voltage is stepped down. The zero
crossing of the input voltage is detected by the ZCD. ZCD output is given to the processor
through the capture unit of Micro- 2812. In the processor sine PWM signal is generated. This
signal is compared with the ZCD signal and a PWM signal is generated. This generated
PWM is applied to the IGBT gate terminal. The PWM signal width is controlled by the I/O
port key.
210
Fig. 6.53. Matrix Converter connection diagram.
Input supply is given from an auto-transformer and is fixed at 100 V rms, 50 Hz. The
switching frequency of the bi-directional power switch of the Matrix Converter is fixed at 8
kHz. The developed Matrix Converter is tested for wide range of output frequencies for as
low as 5 Hz fundamental. Multi-phase R-L load (R = 10 Ώ, L = 10 mH) is connected at the
output terminals of the Matrix Converter. The voltage probe is used with voltage scaling of
1/20. The current is measured with a scaling factor of 2A/V. Horizontal scale used for
measuring the output waveform is 10 msec/div. Source side current waveform is measured
with a horizontal scale of 5 msec/div. The resulting output waveform for fundamental
frequency of 25 Hz and five-phase operation is presented fig. 6.54. Input current results with
both linear and over modulation strategy are shown simultaneously in the fig. 6.55. The
output voltage and current offers good results with sinusoidal waveform. This proves the
viability of the proposed modulation scheme for three to five-phase Matrix Converter.
a.
211
b.
Fig. 6.54. Output side five-phase waveform for 25 Hz; a. Output filtered phase voltages (20V/div) b. Output currents (4A/div). Time scale is 10 ms/div.
Fig. 6.55. Input current using linear modulation scheme as well as input side over modulation strategy (10A/div). Time scale 10 msec/div.
Fig.6.56. Output side waveform for 25 Hz using input side over-modulation strategy; Output currents (4A/div). Time scale 10 msec/div.
6.9b Experimental Investigation of two motor supply
Same experimental setup is utilised for three-phase to five-phase two motor supply, results
obtained are reported in Figs. 6.57 -6.58. The Matrix Converter is connected to a five-phase
R-L load and input voltage of 200 V is supplied. The output of the Matrix Converter contains
two frequency components, one with 50 Hz and one with 25 Hz. One output voltage
frequency component is intended to control one machine and the second frequency
component controls the second machine. The experimental results is intended to show the
212
successful generation of two decoupled frequency component at the output of the Matrix
Converter. The obtained results matches very closely to simulation results.
Fig. 6.57. Experimental results: tope trace, phase Voltage, middle trace adjacent line Voltage, and the bottom trace, non-adjacent line Voltage (y-axis: 200V/div, x-axis: 10
msec/div)
Fig.6.58. Experimental results: output currents, (y-axis 4 amp/div, x-axis: 10 msec/div)
6.9c Experimental investigation on 3x6 phase matrixconverter
Input supply is given from an autotransformer and is fixed at 100 V rms, 50 Hz. The
switching frequency of the bi-directional power switch of the matrix converter is fixed at 6
kHz. The developed Matrix Converter is tested for wide range of output frequencies for as
low as 5 Hz fundamental. Six-phase R-L load is connected at the output terminals of the
matrix converter. The modulation code uses harmonic injection method. The resulting input
and output side waveforms for output fundamental frequency of 25 Hz are presented in
Fig.6.59 and 6.60, respectively. The top trace (Fig. 6.59a) shows all the three applied voltages
and one phase converter input side current. To further illustrate the unity input power factor,
phase ‘a’ voltage and phase ‘a’ currents are depicted in Fig. 6.59b. The current drawn from
213
the three-phase grid is completely sinusoidal as illustrated in Fig. 6.59c which shows
converter side and source side current in one trace.
The unfiltered output phase voltages of four different phases are presented in Fig. 6.60a and
filtered output voltages are depicted in Fig. 6.60b. The output filtered phase voltages and load
currents are shown in Fig. 6.60c. The output voltage and current offers good results with
sinusoidal waveform. This proves the viability of the proposed modulation scheme for three
to quasi six-phase matrixconverter.
[40 V/div. and 10 A/div.]
a.
[40 V/div. and 10 A/div.]
b.
214
[10 A/div]
c.
Fig. 6.59. Experimental results of a 3 to quasi 6 phase matrix converter, a. input side voltages and current, b. phase ‘a’ voltage and phase ‘a’ converter side current, c. converter side and
source side currents. Time scale (10 msec/div).
[150 V/div.]
a.
[40 V/div.]
b.
215
[40 V/div. and 10 A/div.]
c.
Fig. 6.60. Output side six-phase waveform for 25 Hz; a. Output unfiltered phase voltage, b. Output filtered phase voltage, c. output filtered voltage and currents. (10 msec/div).
6.10 Summary
This chapter focused on the development of simple control algorithms for multi-phase multi-motor drive system. Carrier-based sinusoidal PWM control is a generic method in a voltage source inverter. However, the same method when applied to a Matrix Converter need appropriate modification. This chapter elaborated the carrier-based PWM for a three-phase input to five-phase matrix converter supplying a five-phase series-connected/parallel-connected two-motor drive. Further carrier-based PWM for a three-phase input to six-phase output Matrix Converter supplying a series-connected/parallel-connected six-phase machine and three-phase machine is discussed. Additionally carrier-based PWM for a three-phase input to seven-phase output Matrix Converter supplying series-connected/parallel-connected three-motor drive system is discussed. It is seen that the major shortcoming is the lower output voltage of matrix converters. Since the overall output voltage is small, the series-connected and parallel-connected machines will be able to run at lower speeds only.
References
[6.1] Young-Doo Yoon, Seung-Ki Sul, “Carrier-based modulation technique for Matrix Converter,” IEEE Trans. Power Elect., vol. 21, no. 6, pp. 1691–1703, November 2006.
[6.2] Poh C. Loh, R. Rong, F. Blaabjerg, P. Wang, “Digital carrier Modulation and Sampling Issues of Matrix Converter,” IEEE Trans. On Power Elect., vol. 24, no. 7, July 2009.
[6.3] T. Satish, K.K. Mohapatra, N. Mohan, “Carrier-based control of matrixconverter in Linear and over-modulation modes,” Proc. of the 2007 summer computer simulation conf., San Diago, California, pp. 98-105, 2007.
216
[6.4] SK. M. Ahmed, A. Iqbal, H. Abu-Rub, “Carrier based PWM techniques for a three-to-five phase Matrix Converter,” PCIM 2010, 4-6 May 2010,Nuremberg, Germany, pp. 998-1003.
[6.5] A. Iqbal, “Modeling and Control of Five-phase and Six-phase Series-connected Two-motor drive system,” PhD Thesis, Liverpool John Moores University, UK, 2006.
[6.6] A. Iqbal, E. Levi, M. Jones, S.N. Vukosavic, “Generalised Sinusoidal PWM with harmonic injection for multi-phase VSIs,” IEEE 37th Power Electronics Specialist conf. (PESC) Jeju, Korea, 18-22 June 2006, CD_ROM paper No. ThB2-3, pp. 2871-2877.
[6.7] M. Jones, S.N. Vukosavic, E. Levi, A. Iqbal, “A six-phase series connected two-motor drive with decoupled dynamic control,” IEEE Trans. On Industry Application, vol. 41, no. 4, pp. 1056-1066, July/Aug, 2005.
[6.8] M. Jones, E. Levi, S.N. Vukosavic, H.A. Toliyat, “Independent vector control of a seven-phase three-motor drive system supplied by a single voltage source inverter,” Proc. 34th IEEE PESC, vol. 4, pp. 1865-1870, 2003.
217
Chapter 7 Direct Duty Ratio Based Pulse Width
Modulation of Multi-phase Matrix Converter
7.1 Introduction
In this chapter, a PWM strategy based on direct duty ratio calculation approach is presented.
At first generalized direct duty ratio based PWM (DDPM) is developed for three-phase to n-
phase Matrix Converter. This approach does not rely on the discontinuous offset rather a
continuous triangular carrier waveform is employed for the purpose of generating the
switching signals. When the desired output phase voltage references are given, each of them
can be synthesized by utilizing input phase voltages based on per-phase output concept. In
addition, by changing simply the slope of the carrier, the input power factor can also be
controlled, maintaining the sinusoidal input currents. The output voltage is limited to 78.86%
of the input voltage magnitude in case of a three-phase input and five-phase output
configuration. Theoretically this is the maximum output magnitude that can be obtained in
this Matrix Converter configuration in the linear modulation region. Analytical approach is
used to develop and analyse the proposed modulation techniques and are further supported by
simulation results. The major aim of the modulation is to produce two fundamental frequency
output from the Matrix Converter that can be used to control two series/parallel connected
five-phase machines.
Major advantages of the presented scheme (direct duty ratio based PWM) are that it is highly
intuitive and flexible. The output voltage limit is reached by simply adding a third harmonic
component corresponding to the input frequency and the nth harmonic corresponding to the
output frequency with n number of output phase, into the output voltage references. The
output voltage limit has different values for different output number of phases. The proposed
control algorithm is modular in nature and can be extended to any number of input and output
phases. The presented scheme can be employed effectively in variable speed multi-phase
motor drive applications.
218
7.2 Direct Duty Ratio based PWM Technique for a Three-phase to n-phase Matrix Converter for Single-motor drive system
In this section the PWM is discussed based on duty ratio calculation in conjunction with the
generalized three-to n-phase topology of a Matrix Converter. The duty ratio based PWM is
developed by using the concept of per-phase output average over one switching period as
shown in reference [7.1].The developed scheme is modular in nature and therefore, it is
applicable to the generalized converter circuit topology.
A switching period Ts of the carrier wave consists of two sub-intervals, T1 (rising slope of the
triangular carrier) and T2 (falling slope of the triangular carrier). When the carrier changes
from zero to the peak value, the sub-interval is called T1; however, when the carrier changes
from peak to the zero value it is termed as sub-interval T2. The input three-phase sinusoidal
waveform can assume different values at different instants of times. The maximum among
the three input signals is termed as Max, the medium amplitude among three input signals is
termed as Mid, and the smallest magnitude is represented as Min. During interval T1 (positive
slope of the carrier), the line-to-line voltage between Max and Min (
,,,, CBACBA vvvMinvvvMax ) phases is used for the calculation of duty ratio. No
consideration is given to the medium amplitude of input signal. The output voltage should
initially follow the Max signal of the input and then should follow the Min signal of the input.
During interval T2, the two line voltages between Max and Mid ( ,,,, CBACBA vvvMidvvvMax
) and Mid and Min ( ,,,, CBACBA vvvMinvvvMid ) are calculated first. The largest among the
two is used for the calculation of the duty ratio. This is done to balance the volt-second
principle. Two different cases can arise in time interval T2 depending upon the relative
magnitude of the input voltages. If Max-Mid >Mid-Min, the output should follow Max for a
certain time period and then follow Mid for a certain time period. This situation is termed as
Case I. Similarly, if (Max-Mid) < (Mid-Min), the output should follow at first Mid of the
input signal and then Min of the input signal: this is termed as Case II. Thus the DPWM
approach uses two out of the three line-to-line input voltages to synthesis output voltages, and
all the three input phases are utilized to conduct current during each switching period. Case I
and II and the generation of gating signals are further elaborated in the next section.
Case-I:
For condition (Max-Mid) (Mid-Min), the generation of gating pattern for the nth output
phase is illustrated in Fig. 7.1 for one switching period. To generate the pattern, at first the
219
duty ratio .......,,,1 cbanDk , is calculated and then compared with high frequency triangular
carrier signal to generate the nth output phase pattern. The gating pattern for the nth leg of the
Matrix Converter is directly derived from the output pattern. The switching pattern is drawn
assuming that the Max is the phase ‘A’ of the input, Mid is the phase ‘B’, and Min is the
phase ‘C’. The switching pattern changes in accordance with the variation in the relative
magnitude of the input phases. The output follows Min of the input signal, if the magnitude of
the duty ratio is more than the magnitude of the carrier and the slope of the carrier is positive.
The output follows the Max of the input signal if the magnitude of the carrier is greater than
the magnitude of the duty ratio irrespective of the slope of the carrier. Finally, the output
tracks Mid if the magnitude of the carrier signal is less than the magnitude of the duty ratio
and the slope of the carrier is negative. Thus, the resulting output phase voltage changes like
Min→Max→Max→Mid. These transition periods are termed as, 321 ,, nnn ttt and 4nt and
these four sub-intervals can be expressed as in reference [7.1]:
snn TDt 11
snn TDt 12 1
snn TDt 11 13 (7.1)
snn TDt 114
4321 nnnns ttttT
where 1nD is the nth phase duty ratio value, when Case I is under consideration and δis
defined by sT
T1 , which refers to the fraction of the slope of the carrier. Now, by using the
volt-second principle of the PWM control, the following equation can be written:
43210
* .,,.,,.,, nCBAnnCBAnCBA
T
onson tvvvMidttvvvMaxtvvvMindtvTvs
(7.2)
Substituting the time intervals expressions from equations (7.1) into (7.2), yields
,,,,,,
,,.,,.11
0
*CBA
CBACBA
CBACBAn
T
ons
on vvvMaxvvvMaxvvvMid
vvvMidvvvMinDdtv
Tv
s
(7.3)
Where Ts is the sampling period, onon vv ,* are the reference and actual average output voltage
of phase ‘n’, respectively and CBA vvv ,, are the input side three-phase voltages. Max, Mid and
220
Min refer to the maximum, medium and minimum values, Dn represents the duty ratio of the
power switch.
The duty ratio is obtained from equation (7.3) as:
,,,,
,, *
1CBACBA
onCBAn vvvMinvvvMid
vvvvMaxD
(7.4)
where ,,,, CBACBA vvvMidvvvMax
Similarly, the duty ratios of other output phases can be obtained and can subsequently be
used for implementation of the PWM scheme.
B. Case-II:
Now considering another situation of (Max-Mid) < (Mid-Min). The output and the switching
patterns can be derived once again following the same principle laid down in the previous sub
section. Fig. 7.2 shows the output and switching pattern for the nth output phase. Here once
again a high frequency triangular carrier wave is compared with the duty ratio value, 2nD to
generate the switching pattern. The only difference in this case compared to the previous case
is that the interval when the magnitude of the carrier signal is greater than the magnitude of
the duty ratio and the slope is negative, then, the output should follow Mid instead of Max.
Contrary to Case I, for this situation the output must follow Max of the input. The time
intervals 321 ,, nnn ttt and 4nt are the same as in equation (7.1) and now the output phase
voltage is changed with the sequence of Min→Max→Mid→Min. The volt-second principle is
now applied to derive the equation for the duty ratio. The volt-second principle equation can
be written as;
221
okv
*okv
1kD
sT
1kt 2kt 3kt 4kt
1T 2T Fig. 7.1. Output and Switching pattern for nth phase in the Case I.
32410
* .,,.,,.,, nCBAnCBAnnCBA
T
onson tvvvMidtvvvMaxttvvvMindtvTvs
(7.5)
Now once again substituting the time expression from equation (7.1) into equation (7.5), one
obtains:
,,
,,. ,,.,,.,,
,,.,,12
0
*
CBA
CBACBACBACBA
CBACBAn
T
ons
on
vvvMid
vvvMidvvvMaxvvvMidvvvMid
vvvMaxvvvMinDdtv
Tv
s
(7.6)
The duty ratio can now be obtained as:
,,,,.
,,. *
2CBACBA
okCBAn vvvMinvvvMid
vvvvMidD
(7.7)
The switching signals for the bi-directional power switching devices can be generated by
considering the switching states of Figs. 7.1 and 7.2. Depending upon the output pattern, the
gating signals are derived. If the output pattern of phase “n” is Max (or Mid, Min), then the
output phase “n” is connected to the input phase whose voltage is Max (or Mid, Min). The
pulse width modulation algorithm can be explained by the block diagram given in Fig. 7.3.
The input voltages are at first examined for their relative magnitudes and the phases with
maximum, medium, and minimum values are determined. The information about their
relative magnitudes are given to the next computation block along with the commanded
222
output phase voltages. The computation block either uses equation (7.4) or equation (7.7) to
generate the duty ratios depending upon the relative magnitude of the input voltages. The
duty ratio obtained goes to the PWM block. The PWM block calculates the time sub-interval
using equation (7.1). The gating pattern is then derived accordingly and given to the Matrix
Converter.
okv
*okv 2kD
sT
1kt 2kt 3kt 4kt
1T 2T
Fig. 7.2. Output and Switching pattern for nth phase in the Case II.
223
Max
Mid
Min
Equation
No.
(7.4) or (7.7)
PWM ....
*** ,........,, onoboa vvv
CBA vvv ,,
111 .,,........., nba DDD
222 .,,........., nba DDD
11S12S
13S
1nS
2nS
3nS
OR
Fig. 7.3. General Implementation diagram of 3 to k phase Matrix Converter.
7.3 Direct Duty Ratio Based PWM for 3-phase to 5-phase Matrix Converter
A specific case of three-to-five phase Matrix Converter is consodered and DDPWM is
developed. This topology was presented in reference [7.2] where simple carrier based PWM
technique were used and simulation results were obtained. Section (7.3) describes the use of
direct duty ratio PWM for the same topology. The input to the Matrix Converter is three-
phase supply from the grid and the output will be five-phase with varaible voltage and
variable frequency. This type of supply is required for five-phase machine drives. At first the
output pattern is elaborated for one sampling interval. For a particular switching period the
output pattern of phases a,b,c,d & e, are shown in Figs. 7.4 to 7.8, considering only Case I.
Since Case II is similar to Case I with only minor modification, the output pattern and
subsequently the switching pattern can also be derived; however it will not be shown here.
The switching pattern is direclty related to the output patterns as described in section III and
are thus not shown in the Figures 7.4 to 7.8.
The major advantage of the proposed PWM control is it is modular in nature; hence each
phase of the outputs can be modulated separately to follow their references. Depending upon
how the reference or target output voltages are created, two methods are evolved. One
method is termed as ‘without harmonic injection’ and the other is called ‘with harmonic
injection’. The maximum output voltage reaches half of the input voltage if simple sinusoidal
reference voltages are assumed. It is shown in reference [7.3] that the magnitude of the output
voltages can be enhanced by subtracting the common mode third harmonic of the input phase
224
voltages from the input phase voltages. The optimum value of the injected harmonic is
obtained as one fourth of the input maximum magnitude. Thus by adopting the harmonic
injection scheme, the output voltage magnitude reaches 0.75 of the input voltage value.
Further enhancement in the output voltage is achieved by injecting the third harmonic of the
output frequency in the reference output voltage. Thus by injecting one sixth of the
magnitude of the third harmonic, the voltage transfer ratio goes up to 0.866 in the case of a
three-to-three phase Matrix Converter. This is a 15.5% increase compared to the harmonic
injection in the input side voltage only. It is important to note that the value 15.5% is same as
the amount of enhancement of the modulation index in the case of a three-phase voltage
source inverter, which is achieved by harmonic injection when compared to simple carrier-
based scheme.
oav
*oav
1aD
sT
1at 2at 3at 4at
1T 2T
Fig. 7.4. Output pattern of phase ‘a’.
obv
*obv
1bD
sT
1bt 2bt 3bt 4bt
1T 2T
Fig. 7.5. Output pattern of Phase ‘b’.
225
ocv
*ocv
1cD
sT
1ct 2ct 3ct 4ct
1T 2T
Fig. 7.6. Output pattern of phase ‘c’.
odv
*odv
1dD
sT
1dt 2dt 3dt 4dt
1T 2T
Fig. 7.7. Output pattern of phase ‘d’.
oev
*oev
1eD
sT
1et 2et 3et 4et
1T 2T
Fig. 7.8. Output pattern of phase ‘e’.
226
In the case of multi-phase voltage source inverter, a similar concept of the nth harmonic
injection was proposed in reference [7.4] for enhancement of the modulation index. It was
shown that by injecting nth harmonic of magnitude nn
MM n /2
sin1
, where n is the
number of phases, the output voltage increases by
n2cos/1
. A similar approach is thus
used in this thesis to enhance the output voltage magnitude as discussed in the section (7.4).
7.4 Simulation Results for Five-phase Single-motor Drive
Modulation without harmonic injection
Simulation is carried out to study the operation of a three to five-phase Matrix Converter
using the modulation scheme that is elaborated in the section (7.3) known as direct duty ratio
PWM [7.5]. The simulation model is developed in the MATLAB/SIMULINK platform by
using ‘simpowersystem’ block set library.
The output reference voltages are given as:
52cos
5
2
54cos
5
2
54cos
5
2
52cos
5
2
cos5
2
*
*
*
*
*
tqVv
tqVv
tqVv
tqVv
tqVv
ormsinoe
ormsinod
ormsinoc
ormsinob
ormsinoa
(7.8)
where the factor q<=0.5 and Vin-rms are the input phase voltage RMS value. The
implementation block diagram of the proposed PWM scheme is illustrated in Fig. 7.9.
227
*oav
*obv
*ocv
*odv
*oev
axD
bxD
cxD
dxD
exD
*Av
*Bv
*Cv
Fig. 7.9. DPWM implementation block without harmonic injection.
The output target voltage and the maximum, minimum, and medium of the input are given to
the duty ratio calculation block. The duty ratios thus generated 2 1,,,,, orxDDDDD exdxcxbxax
are compared with the high frequency carrier signal. The output pattern is then generated as
per the control law discussed in the previous section. The gate drive signals are then derived
directly from the output voltage pattern. The maximum output voltage is limited to half the
input voltage value (in this case as no common mode voltages are added to the input or
output references). The simulation is done for the whole range of operation of the Matrix
Converter and is tested for wide range of output frequencies. The proposed modulation
scheme and the Matrix Converter topology offer excellent performance with a completely
sinusoidal output and minimal simulation effort. A sample result is presented here for input
frequency of 50 Hz, the output frequency is kept at 25 Hz and the modulation index is kept at
a maximum of 0.5. The switching frequency of the Matrix Converter is kept at 10 kHz and
the simulation step size is fixed at 10 µsec. The input voltage magnitude is fixed at 100 V and
a R-L load is connected at the output of the Matrix Converter with R= 10 Ω and L = 1 mH.
The modulation code is written using the Embedded MATLAB function. A suitable discrete
filter is designed to be used in the simulation. The resulting waveforms are depicted for input
quantities and output quantities in Figs. 7.10 and 7.11, respectively. Output and input
voltages are also depicted on the same plot in Fig. 7.12 to show the maximum modulation
index. The output voltage shows a 25 Hz signal and the input voltage is 50 Hz with output
being half the magnitude compared to the input voltage.
It is seen from Fig. 7.10a that the input is controlled at unity p.f. and is independent of the
output power factor. The input current spectrum is completely sinusoidal with magnitude
228
equal to 4.3 Amp, as seen from Fig. 7.10b and from the locus of the α-β axis input current of
Fig. 7.10c. The total harmonic distortion (THD) in the input current is calculated for up to the
20th harmonic and is found to be only 3.78%, which is well under the prescribed limit of
IEEE 519-1999 standard.
The output voltage waveform of Fig. 7.11a shows the successful conversion of three-phase
input to the five-phase output voltages. The spectrum of filtered voltage (Fig. 7.11b) clearly
indicates the sinusoidal output with a total harmonic distortion of 1.25%, which is well under
the prescribed limit. The THD is once again calculated for low order harmonics upto 20. The
locus of the transformed currents shows that the current component in x-y axis is minimal and
the entire current remains in the α-β axis (Figs. 7.11c and 7.11d). This shows the
effectiveness of the modulation technique since it is successfully eliminating the unwanted x-
y components.
Modulation with harmonic injection
The output voltage is limited to half of the input voltage. Thus this section develops a scheme
to enhance the output voltage magnitude by injecting common mode third and fifth harmonic
into the output voltage references. Third harmonic corresponds to the input voltage frequency
while the fifth harmonic voltages correspond to the output voltage reference frequency. The
optimum amount of third harmonic and fifth harmonic are obtained as rmsinV 12
6 and
rmsinqV 5.48
6 , respectively. The input voltage magnitude is assumed fixed and the output is
variable, hence the output term is multiplied with the modulation index term q, where
7886.00 q .
a.
0 0.02 0.04 0.06-100
-50
0
50
100
Time (s)
Sou
rce
Vol
tage
and
Cur
rent
(V
,A) Voltage
Current
229
b.
c.
Fig. 7.10. Input side waveforms of 3 to 5-phase Matrix Converter: a. Input voltage and input filtered current b. Spectrum input current, c. Input phase current locus.
a.
0 0.01 0.02 0.03 0.04 0.05 0.06-20
0
20
Inpu
t Cur
rent
(A
)
Time (s)
101 102 103 1040
5C
urre
nt S
pect
rum
(A
)
Frequency (Hz)
-5 0 5-5
0
5
is
(A)
i s (
A)
0 0.01 0.02 0.03 0.04-60
-40
-20
0
20
40
60
Time (s)
Out
put F
ilter
ed P
hase
vol
tage
s (V
)
230
b.
c.
d.
0 0.02 0.04 0.06 0.08 0.1 0.12-50
0
50
Out
put V
olta
ge (
V)
Time (s)
101
102
103
1040
20
40
60
Vol
tage
Spe
ctru
m (
V)
Frequency (Hz)
-60 -40 -20 0 20 40 60-60
-40
-20
0
20
40
60
Vo
(V)
V o
(V
)
-50 0 50-50
0
50
-20
20
Vxo
(V)
Vyo
(V
)
231
e.
Fig. 7.11. Output side waveforms of 3 to 5-phase Matrix Converter: a. Five-phase output filtered phase voltages b. Spectrum output filtered voltage, c. locus of α-β axis output voltage,
d. locus of x-y axis output voltage, and e. output voltage and current phase ‘a’.
Fig. 7.12. Input and output voltage waveforms for a 3 to 5-phase Matrix Converter.
It is observed that by injecting only the 3rd harmonic, the output voltage becomes 0.75 of the
input voltage. This increase is same as the one achieved in the three to three-phase Matrix
Converter. Now in the case of three to five-phase Matrix Converter, the 3rd harmonic of
output cannot be injected, hence the 5th harmonic of the output frequency is injected. The
output voltage magnitude thus reaches 0.7886 of the input voltage magnitude by injecting
both 3rd and 5th harmonics. Hence, the overall gain in the output is 5.15%. It is to be noted
here that the same amount of enhancement is achieved by the 5th harmonic injection in a five-
phase voltage source inverter [7.4]. The output voltage reference is given as:
0 0.02 0.04 0.06 0.08-60
-40
-20
0
20
40
60
Time(s)
Out
put v
olta
ge a
nd f
ilter
ed c
urre
nt p
hase
'a' (
V, A
)
Current
Voltage
0 0.01 0.02 0.03 0.04-100
-50
0
50
100
Time(s)
Inpu
t and
Out
put V
olta
ges
(V)
Output voltage
Input voltage
232
ttqVv
ttqVv
ttqVv
ttqVv
ttqVv
ormsinoe
ormsinod
ormsinoc
ormsinob
ormsinoa
52cos
5
2
54cos
5
2
54cos
5
2
52cos
5
2
cos5
2
*
*
*
*
*
(7.9)
The output reference is now the sum of the fundamental and the 3rd and 5th harmonic
components, where the sinusoidal output references ride on a common mode voltage t ,
which is:
tqVtVt ormsinirmsin 5cos5.48
63cos
12
6 (7.10)
The oi , are the input and output frequencies, respectively. The implementation block
diagram of the proposed method of the PWM is shown in Fig. 7.13. The input and output
references, along with their common mode voltages and the relative magnitude of the input
voltages, are fed to the ‘duty ratio calculation' block.
The outputs 2 1,,,,, orxDDDDD exdxcxbxax are compared with a high frequency carrier signal
to generate the output voltage pattern. The gate signals are then derived directly from the
generated output voltage patterns.
The simulation is carried out for the reference voltage generation using harmonic injection.
The parameters used in the model are: input frequency = 50 Hz; output frequency = 25 Hz;
Load resistance=10 Ω; Load inductance=1 mH; Filter inductance=500 µH; Filter
Capacitance=50 µF; Carrier frequency=10 KHz; 100rmsinV V V ‐ 100V. The
simulation results are provided for the maximum input to output voltage ratio of 78.8%. The
resulting input and output waveforms are illustrated in Figs. 7.14 and 7.15, respectively.
Output and input voltages are also depicted on the same plot shown in Fig. 7.16 to show the
maximum achievable output voltage.
233
*oav
*A v*Bv
*Cv
axD
bxD
cxD
dxD
exD
*obv
*ocv*odv
*oev
Fig. 7.13. Modulation Implementation block using harmonic injection.
The unity power factor of the input current is evident from the Fig. 7.14a, and this can be
controlled independently of the load power factor. The input current remains sinusoidal with
magnitude of 10.05 Amp, as seen from Figs. 7.14b and 7.14c. The THD in the input current
is limited to 1.08% and is slightly better than the previous case without harmonic injection in
the target output. The input current is slightly increased when compared to the previous case
due to the higher output voltage of the Matrix Converter and is reflected at the input side as
well.
The output voltage magnitude is increased when compared to output of modulation without
harmonic injection, preserving its sinusoidal nature. The output voltage spectrum is
completely sinusoidal. The locus of the output current remains entirely on the α-β axis and
the x-y components are almost negligible. This proves the effectiveness of the modulation
scheme in eliminating the unwanted low-order harmonics. Once again it is seen that the
proposed Matrix Converter is capable of producing the output of any frequency and
magnitude from zero to 78.86% of the input.
234
a.
b.
c.
Fig. 7.14. Input side waveforms of 3 to 5-phase Matrix Converter: a. Input voltage and current, b. Spectrum input current, c. Input phase current locus.
0 0.02 0.04 0.06-100
-50
0
50
100
Time (s)
Sou
rce
volta
ge a
nd c
urre
nt (
V,A
)
Current
Voltage
0 0.01 0.02 0.03 0.04 0.05 0.06-20
0
20
Inpu
t Cur
rent
(A
)
Time (s)
101
102
103
1040
5
10
15
Cur
rent
Spe
ctru
m (
A)
Frequency (Hz)
-15 -10 -5 0 5 10 15-15
-10
-5
0
5
10
15
is
(A)
i s (
A)
235
a.
b.
c.
0 0.01 0.02 0.03 0.04-100
-50
0
50
100
Time (s)
Out
put f
ilter
ed v
oltg
e (V
)
0 0.02 0.04 0.06 0.08 0.1 0.12-100
0
100
Out
put v
olta
ge (
V)
Time (s)
101
102
103
1040
50
100
Vol
tage
spe
ctru
m (
V)
Frequency (Hz)
-100 -50 0 50 100-100
-50
0
50
100
vo
(V)
v o (
V)
236
d.
e.
Fig. 7.15. Output waveforms of 3 to 5-phase Matrix Converter: a. Five-phase output filtered voltages, b. Spectrum output voltage, c. locus of α-β axis output voltage, d. locus of x-y axis
output voltage and e. Phase ‘a’ voltage and current.
Fig. 7.16. Input and output voltage waveforms for a 3 to 5-phase Matrix Converter with
harmonic injection.
-50 0 50-80
-60
-40
-20
0
20
40
60
80
vxo
(V)
v yo (
V)
0 0.02 0.04 0.06 0.08-100
-50
0
50
100
Time(s)
Out
put v
olta
ge a
nd f
ilte
red
curr
ent p
hase
'a' (
V, A
)
Voltage
Current
0 0.01 0.02 0.03 0.04-100
-50
0
50
100
Time (s)
Inpu
t and
out
put v
olta
ge (
V)
Input voltageOutput voltage
237
7.4 DDPWM of three-to-fivephase Matrix Converters for five-phase two-motor drive
The direct duty ratio based PWM (DDPWM) technique for multi-phase Matrix Converter is
elaborated in the Section 7.2 that is applicable to single-motor drive system (produces one
fundamental frequency component, that can control one machine/load). For controlling two
machines supplied by only one Matrix Converter, the output of the Matrix Converter should
contain two fundamental frequencies.
As discussed in Section 7.2, two different cases can arise in time period T2 depending upon
the relative magnitude of the input voltages. If (Max-Mid) > (Med-Min), the output should
follow Max for a certain time period and then follow Med for ascertain time period. This
situation is termed as Case I. Similarly, if (Max-Med) < (Med-Min), the output should follow
at first Med of the input signal and then Min of the input signal; this is termed as Case II.
Thus the DDPWM approach uses two out of the three line-to-line input voltages to synthesise
output voltages, and all the three input phases are utilized to conduct current during each
switching period.
The duty ratio obtained for case I is:
,,,,
,, *
1CBACBA
okCBAk vvvMinvvvMed
vvvvMaxD
(7.11)
where ,,,, CBACBA vvvMedvvvMax
The duty ratio obtained for case II is:
,,,,.
,,. *
2CBACBA
okCBAK vvvMinvvvMed
vvvvMedD
(7.12)
Depending upon the output pattern, the gating signals are derived. If the output pattern of
phase “k” is Max (or Med, Min), then the output phase “k” is connected to the input phase
whose voltage is Max (or Med, Min). The only difference between single-motor drive and
two-motor drive is the generation of reference output voltage *okv . The output voltage
reference in case of two-motor drive is the sum of the individual five-phase voltage
references corresponding to the operating conditions of the two motors and summed
according to the phase transposition rule as discussed in Chapter 3.
238
The PWM algorithm can be explained by the block diagram given in Fig. 7.17.
The input voltages are at first examined for their relative magnitudes and the phases with
maximum, medium, and minimum values are determined. The information about their
relative magnitudes are given in the next computation block along with the commanded
output phase voltages. The computation block generate the duty ratios depending upon the
relative magnitude of the input voltages. The duty ratio obtained goes to the PWM block. The
PWM block calculates the time sub-interval using equation (7.1). The gating pattern is then
derived accordingly and given to the Matrix Converter.
*oe
*ob
*oa v,........,v,v 111
CBA vvv ,,
111 eba D.,,.........D,D
222 eba D.,,.........D,D
11S12S
13S
51S
52S
53S
*oe
*ob
*oa v,........,v,v 222
Fig. 7.17 General Implementation diagram of 3 to 5phase Matrix Converter.
In the case of multi-phase voltage source inverter, a concept of the nth harmonic injection was
proposed in reference [7.4] for enhancement of the modulation index. It was shown that by
injecting nth harmonic of magnitude nn
MM n /2
sin1
, where n is the number of phases,
the output voltage increases by
n2cos/1
. A similar approach is thus used in this thesis to
enhance the output voltage magnitude. Overall block schematic of the PWM for two-motor
drive system is presented in Fig. 7.18.
239
Fig. 7.18. General Block diagram of PWM for two frequency output.
It is observed that by injecting only 3rd harmonic the output becomes 0.75 of the input. Now
in case of three to five-phase Matrix Converter, 3rd harmonic of output cannot be injected
hence 5th harmonic of the output frequency is injected. For two frequency output, the
injection for both input and output frequency should be proportionally distributed among the
two output frequencies term according to v/f control scheme. The output voltage magnitude
reaches 0.7886 of the input voltage magnitude by injecting both 3rd and 5th harmonic. Thus
the overall gain in the output is 5.15% similar to that of in the reference [7.3]. It is to be noted
here that the same amount of enhancement is achieved by 5th harmonic injection in a five-
phase voltage source inverter. The output voltage reference for the two frequency output is
give as:
ttVq
ttVq
v
ttVq
ttVq
v
ttVq
ttVq
v
ttVq
ttVq
v
ttVq
ttVq
v
ormsinormsinoe
ormsinormsinod
ormsinormsinoc
ormsinormsinob
ormsinormsinoa
21*
21*
21*
21*
21*
52*2cos
3
*2
5
2
52cos
35
2
54*2cos
3
*2
5
2
54cos
35
2
54*2cos
3
*2
5
2
54cos
35
2
52*2cos
3
*2
5
2
52cos
35
2
*2cos3
*2
5
2cos
35
2
(7.13) The sinusoidal output references ride on a common mode voltage t1 and t2 , which are:
h t ‐ √ V ‐ . sin 6πf t ‐ √
.. q. V ‐ . sin 10πf t
240
tVq
tVt ormsinirmsin *2*5cos3
*2
5.48
63cos
24
62
(7.14)
The first term of equation (7.14) is related to the input side and the second term is for the
output side.
7.5 Simulation Results for Two-motor drive system
MATLAB/SIMULINK model is developed for the proposed Matrix Converter control. The
input voltage is fixed at 100 V to show the exact gain at the output side. The switching
frequency of the devices is kept at 10 kHz., Load resistance = 4 Ω; Load inductance =1 mH;
Filter inductance = 500 µH; Filter Capacitance = 50 µF.
The input is three phase ac source and the output is five phase. The output is a combination of
two frequencies controlled by the direct duty ratio PWM scheme. The output loads are
connected in series with phase transposition so that the d-q and x-y components of the output
five phases can be utilized. DDPWM scheme needs the reference values of input three phases
and output five phases. Time is given as the third input for the direct duty ratio PWM block
sub system.
The purpose here is to show two decoupled fundamental components of voltages produced by
the Matrix Converter. The load to the Matrix Converter can be two series/parallel-connected
five-phase machines. These voltage components produced by the Matrix Converter are
decoupled from each other and thus can independently control the two machines. The results
shown here is only limited to the production of the appropriate voltage components. The
motor behaviour is not discussed in this thesis. It is assumed that one voltage component has
frequency of 25 Hz (750 rpm of the motor) and second voltage component has 50 Hz (1500
rpm of the motor). To respect the v/f = constant control the voltage magnitude of the lower
frequency component is half compared to the higher frequency component. Simulation results
are shown for the modulation with common mode voltage addition in the output target
voltage to obtain the maximum possible voltage at the output of the Matrix Converter. Thus
the maximum output of the Matrix Converter is limited to 78.8 V as the input is 100 V. Two
voltage references are generated corresponding to the two operating speeds of the motors (25
Hz and 50 Hz in this case) and are added as per the transposition rule. The voltage references
thus generated are the modulating signals. The duty ratios are then calculated and used further
to generate the switching signals for the bi-directional power switches of the Matrix
Converter.
241
The resulting waveforms are presented in Figs. 7.19 to 7.21. The input source side (at the
three-phase supply) and converter side (Matrix Converter input side after the filter)
waveforms are presented in Fig. 7.19. The results clearly show unity power factor at the input
side. The converter side current shows PWM signals while the source side currents are
sinusoidal. Thus the input filter design is satisfactory to eliminate the switching harmonics
from the PWM signals. The lower-order harmonics are not present in the converter side or
source side currents. This is the special feature of the Matrix Converter as the input currents
from the source are sinusoidal.
a.
b.
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-100
-50
0
50
100
Time [s]
Sou
rce
side
vol
tage
and
cur
rent
pha
se 'a
' [V
, A]
Source Voltage
SourceCurrent
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-15
-10
-5
0
5
10
15
Time [s]
Sou
rce
side
cur
rent
s [A
]
242
c.
Fig. 7.19. The input side waveforms; a. source side voltage and current for phase ‘a’, b. source side three-phase currents, c. converter side three-phase currents.
The output side waveforms are depicted in Fig. 7.20a and the spectrums are shown in Fig.
7.20b. Two system of line voltages exist in a five-phase system adjacent line voltages (Vab,
Vbc, Vcd, Vde) and non-adjacent voltages (Vac, Vbd, Vce, Vdb). One phase voltage, Va, one
adjacent line voltage, Vab and one non-adjacent line voltage Vac are depicted in Fig. 7.20a.
The filtered five-phase voltages are presented in Fig. 7.20b, that shows the combination of
fundamental and its second harmonic voltages. The output inverter currents are presented in
Fig. 7.20c that also shows two components.
a.
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-50
0
50
I a [A
]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-50
0
50
I b [A
]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-50
0
50
I c [A
]
Time [s]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-200
0
200
Vph
ase [
V]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-200
0
200
Vad
j-lin
e [V
]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-200
0
200
Vno
n-ad
j-lin
e [V
]
Time [s]
243
b.
c.
Fig. 7.20. The output side waveform; a. Phase, adjacent and non-adjacent voltages, b. filtered output phase voltages, c. five-phase inverter currents.
The output filtered phase voltage of the Matrix Converter is transformed in the stationary
reference frame into two orthogonal components namely V and Vxy and are shown in Fig.
7.21. The V components control one machine which operates at 25 Hz and the other
components Vxy control the other machine that operate at 50 Hz. The peak voltage of the V
component is half that of the Vxy component as the frequency of operation is half and thus v/f
= constant principle is maintained. It is evident that both the componens of voltages are
independent and sinuosoidal.
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-80
-60
-40
-20
0
20
40
60
80
Time [s]
Out
put
filt
ered
vol
tage
s [V
]
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-20
-15
-10
-5
0
5
10
15
20
Time [s]
Out
put
Inve
rter
Cur
rent
[A
]
244
a.
b.
Fig. 7.21. The transformed voltage of the output phase voltages: a. and b. Vxy.
The FFT analyis is further carried out for the phase voltage and the transformed voltages to
prove the decoupled components of the voltages. The resulting time domain and frequency
domain voltages are depicted in Fig. 7.22. The phase ‘a’ voltage shows two component of
voltages at the fundamental frequencies of 25 Hz and 50 Hz. The 25 Hz component is taken
as the first fundamental voltage and has 100% magnitude of 25.9 V, while the 50 Hz
component shows the magnitude of 200% i.e. 25.9 x 2 = 51.8 V. The sum of these two
voltages are (51.8 + 25.9 = 77.7 V) which is slighlty smaller than the therotecial value of 78.7
V and this is due to the numerical error in the computation and the sampling time of the
simulation model. The transformed voltage are further shown in Figs. 7.22b and 7.22c and
these two components shows peaks at 25 Hz and 50 Hz. Hence, it is evident that the two
components are independent of each other and will subsequently control the two machines
independently. Thus the proposed PWM control of the Matrix Converter succefully produces
two independent voltage components.
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-30
-20
-10
0
10
20
30
Time [s]
V [
V]
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-60
-40
-20
0
20
40
60
Time [s]
Vxy
[V
]
245
a.
b.
c.
Fig. 7.22. Time domain and frequency domain waveforms: a. Phase ‘a’ voltage, b. voltage and c. Vx voltage.
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-100
0
100
Selected signal: 2 cycles. FFT window (in red): 2 cycles
Time [s]
Va [
V]
101
102
103
104
0
50
100
150
200
Frequency [Hz]
Fundamental (25Hz) = 25.9 V
Spe
ctru
m [
%]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-50
0
50
Selected signal: 2 cycles. FFT window (in red): 2 cycles
Time [s]
V
[V
]
101
102
103
104
0
50
100
Frequency [Hz]
Fundamental (25Hz) = 25.87
Spe
ctru
m [
%]
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-100
0
100Selected signal: 4 cycles. FFT window (in red): 2 cycles
Time [s]
Vx [
V]
101
102
103
104
0
50
100
Frequency [Hz]
Fundamental (50Hz) = 51.23 V
Spe
ctru
m [
%]
246
The output side filtered voltages are given in Fig. 7.19 and the spectrum of the output PWM
signal voltages are presented in Fig.7.21. The output filtered phase voltages shows
superimposed fundamental and second harmonic components. The spectrum of phase ‘A’
voltage and the transformed voltage are given in Fig. 7.21. It is clearly evident that the phase
‘A’ voltage contains two fundamental components at 25 Hz and 50 Hz. These voltages are
then decoupled and appear in α-β plane (50 Hz) and x-y plane (25 Hz). Thus the aim of the
control is achieved. Also the magnitude of the two voltages follows v/f = constant rule.
7.6 Experimental Results
The Matrix Converter is connected to a five-phase R-L load and input voltage of 200 V is
supplied. The output of the Matrix Converter contains two frequency components, one with
50 Hz and one with 25 Hz. One output voltage frequency component is intended to control
one machine and the second frequency component will control the second machine. The
experimental results is intended to show the successful generation of two decoupled
frequency component at the output of the Matrix Converter. The obtained results matches
very closely the simulation results. The experimental results are shown in figs.7.23-7.24
Fig. 7.23a . Experimental results: tope trace, phase Voltage, middle trace adjacent line Voltage, and the bottom trace, non-adjacent line Voltage (y-axis: 200V/div, x-axis: 10
msec/div).
247
Fig. 7.23b. Experimental results: output currents, (y-axis 4 amp/div, x-axis: 10 msec/div)
7.7 Summary
This chapter present direct duty ratio based PWM for a three-phase to five-phase Matrix Converter supplying series-connected/parallel-connected five-phase two-motor drive system. The presented PWM is different from carrier-based and space vector methods. The presented scheme is modular in nature and can be applied to any number of output phases of Matrix Converter. The output limit is same as that of obtainable using carrier-based and space vector PWMs. However, the simplicity of approach is the major attraction of direct duty ratio based PWM methods.
References
[7.1] Yulong Li, Nam-Sup Choi, Byung-Moon Han, Kyung Min Kim, Buhm Lee and Jun-Hyub Park, “Direct duty ratio pulse width modulation method for Matrix Converter,” Int. Journal of Control Automation and Systems, vol. 6, no. 5, pp. 660-669, Oct. 2008.
[7.2] A. Iqbal, Sk. M. Ahmad, H. Abu-Rub and M.R. Khan, “Carrier based PWM technique for a novel three-to-five phase Matrix Converter,” Proc. European PCIM, CD-Rom paper no. 173, 2-6 May Nuremberg, Germany, 2010.
[7.3] M.P. Kazmierkowaski, R. Krishnan and F. Blaabjerg, “Control in Power Electronics-Selected problems,” Academic Press, USA, 2002.
[7.4] A. Iqbal, E. Levi, M. Jones, S.N. Vukosavic, “Generalised Sinusoidal PWM with harmonic injection for multi-phase VSIs,” IEEE 37th Power Electronics Specialist conf. (PESC) Jeju, Korea, 2006, CD_ROM paper No. ThB2-3, pp. 2871-2877, 2006.
[7.5] B. Wang, G. Venkataramanan, “A carrier-based PWM algorithm for indirect Matrix Converters,” in Proc. IEEE-PESC 2006, pp. 2780–2787, 2006.
248
Chapter 8 Conclusions and Future Work
8.1 Conclusions
Research on variable speed multi-phase motor drive systems has seen a significant growth in
recent years. The main driving force behind the accelerated research input is the advent of
cheap and reliable semiconductor power switching devices and fast and efficient Digital
Signal Processors/Field programmable gate arrays/Microcontroller/Embedded controllers etc.
Thus more complex control algorithm can be implemented in real time more efficiently. The
development in power electronic converter technology has also enabled the number of phases
to be considered as an additional design parameter. Multi-phase (more than three phases)
machines offer some major advantages over three-phase ones, such as reduced torque
pulsation and pulsation at higher frequency, higher fault tolerance, lower per-phase power
handling requirements, enhanced modularity, lower DC link harmonics and improved noise
characteristics. In spite, notable advantages of multi-phase machines, ‘off the shelf
availability’ of three-phase machines still limits the application of multi-phase machine to
specialized applications, for which three-phase drives are either not readily available or do
not satisfy the specification. One of the important application areas due to the high fault
tolerance of multi-phase drives, is in aerospace, predominantly in conjunction with multi-
phase Permanent Magnet machines. Applications of multi-phase machines in electric and
hybrid electric vehicles for propulsion and power steering are also feasible. Ship propulsion is
considered as one of the main areas of application formulation-phase drives. Multi-phase
Wind turbine generators are also considered for remote offshore applications especially using
six-phase configuration. The multi-phase output generated by wind generator is rectified to
DC and the power is transferred to shore using HVDC.
Multi-motor drive system (series-connected/parallel-connected) supplied by a single power
converter is an attractive features of multi-phase system. Stator windings of two or more
machines (depending upon the number of phases) can be connected in series/parallel and
supplied by one variable voltage and variable frequency source and controlled using field
oriented control principle, can independently control the machines. The major advantage of
249
this drive configuration is the saving in the number of converter legs. The major application
areas for multi-motor drive system could be in ship propulsion (main propeller and auxiliary
functions), mining (due to space restriction), winder application (two machines runs to wind
and unwind finished product) etc.
The multi-phase multi-motor drive system is readily supplied by current controlled pulse
width modulated inverters. In this thesis, alternative power supply solution is explored and a
direct Matrix Converter is considered.
This thesis is thus aimed at development of modeling and control algorithms for multi-phase
direct Matrix Converter. The Matrix Converter topology investigated in this thesis has three-
phase input and multi-phase output. The most common power converter supplying a multi-
phase motor drive for variable speed applications is a voltage source inverter. However, the
major drawback of a voltage source inverter is the presence of bulky DC link capacitors.
Further, the quality of source side current (utility grid side) in terms of high total harmonic
distortion is poor, due to presence of diode based rectifier at the front end. Moreover, the
power factor at the source side in not controllable. The alternative to this topology is voltage
source inverter with active front end rectifier, also called back-to-back converter. This
topology offers controllable source side power factor, bi-directional power flow and
sinusoidal source side (utility grid side) current. However, the DC link capacitor is
unavoidable which add to the cost and volume. An alternative topology is a Matrix
Converter.
In a general case, a Matrix Converter can be viewed as an array of n×k bi-directional power
semiconductor switches that are able to transform n-phase input voltages into k-phase output
voltages of variable magnitude, phase and frequency. The circuit configuration, typically has
no energy storage element and eliminate the need of a DC-link circuit. The input side of the
Matrix Converter is considered as voltage fed, while the output is a current fed system. For
this reason a capacitive filter on the input and an inductive one on the output are necessary.
The size of the filter is significantly lower in comparison to the equivalent voltage source
inverter.
Since space vector approach offers more generic method of modeling, it is adopted in the
thesis for modeling a five-phase, six-phase and a seven-phase direct Matrix Converter. This is
followed by development of various control algorithms. For the considered 3-phase (input) to
k-phase (output) configuration, the theoretical number of switching states amounts to 23•k, out
250
of which some are not useful for implementing modulation algorithms. Considering the
nature of the two sides of the converter, the constraints are that two phases of the input should
never be short circuited and the output current should not be interrupted. As a result, exactly
one switch should conduct in each output phase in every instant. This makes the number of
usable switching states 3k. The complete set of switching states and usable vectors are given
in the thesis.
In case of a voltage source inverter, the output voltage level is fixed such as 2-level, 3-level
etc. However, in case of a Matrix Converter, the number of output voltage level can be of any
value by using any of the values of the voltages of the input 3 phases. The result is a complex
switching logic in a Matrix Converter, however, the availability of desired voltage level at the
output side is a significant advantage of a Matrix Converter. The basic operating principle of
a Matrix Converter, is to piece together an output voltage waveform with the desired
fundamental component from selected segments of the input voltage waveforms. For safe
operation of a Matrix Converter, an adequate clamping circuit is usually added to it that allow
output current free-wheeling path. This clamping circuit usually incorporates a capacitance
which is significantly smaller than the one used in counterpart Voltage Source Inverters.
The multi-phase Matrix Converters (multi-phase refers to three-phase input and more than
three-phase output) use the same bi-directional power semiconductor switching device and
apply the same switching strategies as in the three-phase case. Multi-phase Matrix Converters
represent a significant challenge when developing suitable modulation methods. This
challenge is even more pronounced than in the case of the multi-phase VSIs, due to the
higher number of possible switching states in Matrix Converters.
The modulation strategies considered in this thesis are;
Carrier-based PWM, and
Direct-duty ration based PWM
Carrier-based modulation methods, have been used for three-to-five, three-to-six and three-
to-seven phase topologies in. The duty ratios obtained using analytical method represent
sinusoidal functions, similar to the coefficients of a rotational transformation to the
synchronous reference frame, allowing the resulting voltage to be independent of the three-
phase supply frequency. This is a critical requirement of modulation of a Matrix Converter
using carrier-based scheme. The change of the phase shift and magnitudes of these
251
coefficients results in the alteration of the phase shift and magnitude of the output voltages.
Thus the carrier-based PWM scheme used to independently control multi-phase multi-motor
drive system (five-phase, six-phase and seven-phase) can effectively control the
series/parallel-connected machines. The source side current is sinusoidal with controllable
power factor.
In direct duty ratio based PWM, duty ratios is calculated for each switch, and is compared to
a carrier signal. It applies to each output phase independently and therefore, can be used to
supply multi-phase loads. The input number of phases is restricted to three, since the
modulation requires identification of the values of the input voltages as the highest, medium
and the lowest. Once again independence of control is achieved for a five-phase
series/parallel-connected two-motor drive system. The source side current is sinusoidal with
unity power factor. The output voltage shows two fundamental frequencies that are used to
control two machines independently. This is a modular approach and can be used for any
number of phases. By changing the slope of carrier the power factor of the source can be
varied.
The voltage transfer ratio between the output voltage to input voltage magnitude for different
phase numbers is obtained as (first column given the number of input phase number which is
3 and the second to 5th column shows the output phase number);
Input phase/Output phase 3 5 6 7
3 86.6% 78.8% 75% 76.9%
It is observed that the output voltage magnitude reduces as the number of output phases
increases. In case of series/parallel-connected multi-phase multi-motor drive system, this
voltage has to be appropriately distributed among different machines. Since the output of a
Matrix Converter voltage is lower, supplying more than one machine, is difficult and voltage
reserve issue may arise. Either using lower rated voltage machines or using boost in the
Matrix Converter (AC chopper) is seen as possible solution. However, this is not considered
in the thesis. The major aim of the thesis is to investigate the possibility of generating more
than one independent fundamental frequency components and this is successfully achieved.
252
8.2 Future Work
The work carried out in thesis is limited to the modelling and control of a five-phase, six-
phase and seven-phase inverter with resistive (R) and resistive and inductive (R-L) load in
open-loop mode. However, there is still a huge scope of further research on this topic. The
motor drive supplied using the developed schemes have not yet been taken up. Hence
performance evaluation of multi-phase motors fed using multi-phase Matrix Converter with
the proposed control algorithms could be a direction of further research.
The multi-phase Matrix Converters provide improvements compared to the three-phase
counterparts. In this respect, converters with multi-phase input or output phase number can be
considered separately. When the number of input phases is higher than the output, the likely
applications would be wind generation interface system. This configuration of Matrix
Converter potentially provides higher resolution in the achievement of the output voltage and
also a higher achievable output voltage. There exists huge scope in further research in this
direction.
However, if the Matrix Converters are considered as multi-level inverters with their DC bus
potentials being variable, such a topology can be considered quite promising. This can also be
taken up in future research.
The effect of input side disturbances in terms of voltage variation, waveform distortion and
change in grid frequency, can be taken up for investigation in future work.
Multi-phase multi-level Matrix Converter can also be investigated and this is a good leading
direction of research. Voltage boost using impedance network at the source side or AC
chopper at the output side could be considered as a potential direction of research.